JOURNAL OF BACTERIOLOGY, Feb., 1966 Vol. 91, No. 2Copyright © 1966 American Society for Microbiology Printed in U.S.A.

Light-Induced Lysis and Carotenogenesisin Myxococcus xanthus'

ROBERT P. BURCHARD AND MARTIN DWORKIN2Department of Microbiology, University of Minnesota, Minneapolis, Minnesota

Received for publication 30 July 1965

ABSTRACTBURCHARD, ROBERT P. (University of Minnesota, Minneapolis), AND MARTIN

DWORKIN. Light-induced lysis and carotenogenesis in Myxococcus xanthus. J. Bac-teriol. 91:535-545. 1966.-Myxococcus xanthus, grown vegetatively in the light,developed an orange carotenoid after the cells entered stationary phase of growth;pigment content increased with age. Cells grown in the dark did not develop ca-rotenoid and could be photolysed by relatively low-intensity light only duringstationary phase; rate of photolysis increased with age. Photolysis adhered to thereciprocity law, was temperature-independent and oxygen-dependent, and requiredthe presence of nonspecific, monovalent cations; it was inhibited by one of severaldivalent cations. Logarithmic-phase cells were photosensitized by 100,000 X gpellet preparations of sonic-treated stationary-phase cells grown in the light anddark. A porphyrin with a Soret band at 408 m,u was isolated from photosensitivecells; logarithmic-phase cells contained about 1X6 the amount of porphyrin ofstationary-phase cells. The purified material had spectral and chemical propertiesof protoporphyrin IX and photosensitized logarithmic-phase cells. Its spectrumwas similar to the action spectrum for photolysis. We concluded that protoporphy-rin IX is the natural endogenous photosensitizer. Carotenogenesis was stimulatedby light in the blue-violet region of the visible spectrum and was inhibited by di-phenylamine, resulting in photosensitivity of the cells. Photoprotection by carote-noid was lost in the cold. A mutant which synthesized carotenoid in the lightand dark was photosensitive only after growth in diphenylamine. The ecologicalsignificance of these phenomena is discussed.

The observation which led to this investigationwas that colonies of Myxococcus xanthus FB onagar, incubated for 7 to 8 days in the dark andthen inadvertently exposed to light, lysed fromtheir centers outward to their peripheries. Pre-liminary attempts to attribute this phenomenonto induction of a prophage by visible light wereunsuccessful. Further investigation of the phe-nomenon revealed it to be the result of photolysisof physiologically old cells grown in the dark.It was also observed that colonies developing inthe light produced orange pigment, probablycarotenoid, and were insensitive to light.Blum (2) has described several photodynamic1 Based on a thesis submitted by the senior author

in partial fulfillment of requirements for the Ph.D.degree. Presented in part at the 65th Annual Meetingof the American Society for Microbiology, AtlanticCity, N.J., 25-29 April 1965.

2 Public Health Service Career DevelopmentAwardee 1-K3-GM-5869-101.

diseases of man and higher animals. Among thelower protista, Corynebacterium poinsettiae (14),Sarcina lutea (21), Mycobacterium marinum (19),and Halobacterium salinarium (6) can be photo-sensitized by exogenous dyes when deprived oftheir carotenoids. Sistrom et al. (25) studiedphotodynamic killing of a carotenoidless mutantof Rhodopseudomonas spheroides and concludedthat endogenous bacteriochlorophyll was thephotosensitizer. Among chemotrophic bacteriademonstrated to be photosensitive, neither iso-lation nor identification of a photosensitizer hasbeen reported. As indicated above, carotenoid-containing microorganisms are capable of beingphotosensitized when deprived of their pigment.Griffiths et al. (13) proposed that the primaryfunction of carotenoids is a photoprotective one.M. xanthus is an inhabitant of the soil; this

plus its ability to move over a semisolid substratewould expose the organism to alternating lightand dark conditions. As indicated above, light

535

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

BURCHARD AND DWORKIN

has marked effects on M. xanthus. The role oflight as an inducer of lysis is considered in thispaper specifically in terms of the photolytic proc-ess, conditions required for photosensitivity, andthe location and identity of the photosensitizer.In addition, some data pertaining to light-inducedcarotenogenesis are presented.

MATERIALS AND METHODS

Organism. M. xanthus FB (8), a dispersed-growingmyxobacterium, routinely dissociated into yellow andtan colonies. Ultraviolet irradiation of FB and re-peated selection of a tan variant eventually gave riseto a homogenous nondissociating form (FBt). FBy,a yellow variant, was selected as a spontaneous raremutant of FBt. Both FBt and FB, synthesized ca-rotenoid pigments only in response to visible light(see Results). FBt gave rise spontaneously to anotherstable variant, FBt dc, which synthesized carotenoidsin the dark as well as in the light.

Cultivation of vegetative cells. Cells were culturedin CT-1 liquid medium [2% Casitone (Difco), 10-2 MK2HPO4-KH2PO4 (pH 7.6), and 1.85 X 107- M

MgSO4; this differs slightly from CT (8), in that theMgSO4 concentration was halved] on a rotary orreciprocating shaker at 30 or 22 C, respectively. Cul-tures were incubated in the light under fluorescentlamps (GE, cool white, 15 w) at 500 to 1,000 ft-c, asmeasured by a Photovolt Universal photometer,model 200. Cultures incubated in the dark werewrapped in aluminum foil.

Age ofcells. Logarithmic-phase cells were generallysampled at a culture density of 2 X 108 to 9 X 108 permilliliter. When stationary-phase cells were used, theculture was generally at a concentration of 1.2 X 109to 2.5 X 109 cells per milliliter.

Enumeration of cells. Cell counts were carried outwith a Petroff-Hausser counting chamber and a Zeissphase-contrast microscope. Dilution of cells, when re-quired, was carried out in deionized water. The spread-plate technique on CT-1 agar (incubation at 30 C for5 to 6 days) was used for viable counts.

Cultivation of phenotypically carotenoidless cells.Carotenoid synthesis was inhibited by addition of 10-4M diphenylamine (11) to the growth medium. Cultureflasks were wrapped in red, gelatin film, transmittinglight primarily above 590 mA. This quality light nor-mally induced some carotenogenesis; red light did notphotodecompose diphenylamine (14).

Extraction of pigments. The bacteria were sedi-mented by centrifugation of the culture medium at10,000 X g. Wet pellets were suspended in acetone.The suspension was centrifuged, leaving most of thepigment in the supematant fraction.

Preparation of cell-free extracts. Large quantitiesof cells were harvested by centrifugation, washed twicein 0.02 M phosphate buffer (pH 7), and resuspended ina small but workable volume of buffer (all steps werecarried out at 4 C). The washed cell suspension wassonic-treated for 1.5 min in a 20-kc ultrasonic disin-tegrator (Measuring and Scientific Equipment, Ltd.).The resulting preparation of ruptured cells was cen-trifuged at 10,000 X g for 15 min to eliminate the few

whole cells and large debris. The supernatant fractionwas centrifuged in a model 40 head in a Spinco L-2centrifuge at 4 C for 75 min at 100,000 X g. The pelletwas then resuspended in phosphate buffer (0.02 M,pH 7).

Isolation of porphyrin. A variation on the methodof Bogorad and Granick (3) was employed for ex-traction of porphyrins. Approximately 1.3 X 108'cells (about 1 g, dry weight) were harvested, washedonce in deionized water, and extracted twice with 75ml of glacial acetic acid-concentrated hydrochloricacid (98: 2). The preparation was centrifuged at10,000 X g, and the supernatant fractions were pooledand adjusted to pH 3.5 with concentrated ammoniumhydroxide. Approximately 0.5 volume of water wasadded, and the pigment was then extracted into ether.The ether was extracted with 0.1 and 1.0 N HCl insequence, and the 0.1 N extract was discarded.

Procedure for general photolysis experiments. Cellsuspensions were placed in screw-capped, glass con-tainers to a depth of 1 to 2 mm, and were exposed tocool white, 15-w fluorescent lamps (1,000 to 1,500 ft-c)on a reciprocating shaker at 22 C. When higher-in-tensity light was required, a 75- or 500-w GE reflectorspotlight was used. The latter required the use of afan and the interposition of a glass dish of water be-tween the cells and the light source to prevent over-heating of the suspension. Under these conditions, the500-w lamp supplied approximately 18,000 ft-c, andthe temperature of the cell suspension did not riseabove 30 C.

Preliminary action spectrum. Light-sensitive cellswere placed in screw-capped, square, tablet bottles.Colored filters (gelatin films, Klett-Summersoncolorimeter filters, or a Kodak Wratten 18A filter)were attached to the upper face of the bottles. Theremaining exposed glass was covered with aluminumfoil.

Spectroscopy of cells and subcellular fractions. Dif-ference spectra of 100,000 X g pellets and other frac-tions were carried out according to the method ofDworkin and Niederpruem (9). Most of the differencespectra and pigment extracts were scanned with aBeckman DK-2A spectrophotometer.

RESuLTSGrowth in light and dark. M. xanthus FBt was

cultivated in the light and dark; there was nodetectable difference in growth rate and yieldof the organisms under either growth condition.

Observations on photolysing cells. VV ecells examined by phase-contrast microscopywere long, slender, flexible, optically dense rods.After exposure of photosensitive cells grown inthe dark to light, microscopically visible changesbegan to occur. The cells became vacuolar orgranulated, and distorted. Opacity of an indi-vidual cell decreased until the cell appeared as-aghost. Finally, all structural integrity was lost.With the appearance of visible alterations, thecells were no longer viable. Figure 1 indicatesthe total and viable counts of cultures incubated

536 J. BAamiumO.

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

LYSIS AND CAROTENOGENESIS IN M. XANTHUS

io? L DARK;VOTAL CELLSAP--&DANC; vuLE CELS

4 ~ -o UGMM3 TOTNL CELLI5 fr--AL6"TVIA3 E CELLS

106 I

0 50 100 150 200TIME IN MINUTES

FIG. 1. Total and viable cell counts during photolysisofMyxococcus xanthus FBt .

in the light and in the dark. The Petroff-Haussercounts of visually unaltered cells correlate wellwith viable count and were used as a parameterfor viability.

Appearance of photosensitivity. To determineat what stage of growth the dark cells developedphotosensitivity, a number of flasks were inocu-lated with equal amounts of logarithmic-phasecells and incubated at room temperature. For theentire experiment one flask was incubated in thedark, and one in the light. Each of the otherflasks was illuminated for a 12-hr period andincubated in the dark for the remainder of theexperiment. Data plotted in Fig. 2 indicate thatphotolysis occurred in cultures exposed to lightfor the first time at 36 hr or later. The logarithmicphase of growth ended at approximately 24 hr;carotenoid appeared in the culture exposed con-tinuously to light and in that exposed during the24- to 36-hr period.To determine the relationship between age of

cells in stationary phase and rate of photolysis,FBt cells grown in the dark were exposed to lightat various times after the onset of the stationaryphase (Fig. 3). As the cells aged, the lag in thesurvival curve became shorter, the rate of lysisincreased, and the per cent survivors decreasedsharply.

Characteristics of photolysis. Photolysis gen-erally follows a triphasic curve. When the loga-rithm of the number of the surviving cells wasplotted against time of exposure to light, a lag

-J

Ic~

U)

I-

R 101

It \ ~~~~~~~~~~~~~~~~~II UeTI

r

- DK E

0 12 24 36 48 60 72 1TIME OF INCUBATION (HOURS)

-884

FIG. 2. Effect of light exposure at various initervalsduring the growth cycle of Myxococcus xanthus FBt .

lo1 o0mm

IIJ

W11H0 5.S?

0.

l

31.3&3HRS

0 50 100 150 200 250 300TIME OF LIGHT INCUBATION (MINUTES)

FIG. 3. Photolysis of Myxococcus xanthus FBt atdifferent times (0, 3.5, 8.3, and 31.3 hr) after onset ofthe stationary phase.

occurred before cell death was detectable. Thislag was followed by two phases of exponentialdeath, the second steeper than the first. Withhigher light intensity or more sensitive cells, orboth, we were unable to detect the lag phase or

VOL. 91, 1966 537

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

BURCHARD AND DWORKIN

TABLE 1. Monovalent cationt requirement forphotolysis of Myxococcus xanthus FB,

Salt Concn 2 Incubation inSalt Concn ~~~~~~light*

_- No lysisK2HPO4-KH2PO4...... 1 X 10-2 LysisNa2HPO4-NaH2PO4 1>I<X 10-2 LysisKCI................. 2 X 10-2 LysisNaCl................. 2 X 10-2 LysisLiCl. 2 X 10-2 LysisCaC12.2 X 10-2 No lysisMgSO4 ................ 2 X 10(2 No lysis

* Incubation in the dark consistently showed nolysis.

one phase of exponential death, or both. Thephotolytic process adhered to the Bunsen-Roscoereciprocity law; i.e., it was independent of lightintensity over the range tested (300 to 660 ft-c).It was also independent of temperature (between4 and 22 C; Fig. 9).To determine whether photolysis is oxygen-

dependent, as is characteristic of all photodynamicprocesses (2), photosensitive cells were placed inThunberg tubes, which were evacuated andflushed 10 times with N2. The cells were thenexposed to light in a N2 atmosphere; they didnot photolyse, whereas control cells in air did.

Role of cations in photolysis. Photolysis isnonspecifically dependent on the presence ofmonovalent cations (Table 1). No difference inthe rate of photolysis was detected with K+,Na+, and Li+.To determine whether light and monovalent-

cation dependence are separable, photosensitiveFBt cells suspended in water were exposed tolight for the same length of time required for a100-fold drop in cell number when a water plusK+ suspension of cells was irradiated. The ir-radiated cell suspension was then divided intotwo parts; one was placed in the dark and onewas placed in the light. K+ was added to both,resulting in an almost immediate decrease in theoptical density of both suspensions and distor-tion and vacuolization of the cells; no lysisoccurred. Control cells incubated in water in thedark lysed upon exposure to light and K+; K+added in the dark resulted in no detectable change.The lytic process was inhibited in the presence

of one of several divalent cations. MgCl2, BaCi,or CaCl2 at 10-2 M were equally effective. Cellsgrown in CT (containing twice the MgSO4 con-centration as CT-1) into the stationary phase werenot photosensitive in the growth medium or inwater plus K+.

Location and identification of the photosensi-

tizer. To identify the photosensitizing moiety ofstationary-phase cells of M. xanthus, a prelimi-nary action spectrum for photolysis was carriedout with a series of filters (detailed descriptionin Burchard, Ph.D. Thesis, Univ. Minnesota,1965). The data led to the suggestion by Burchardand Dworkin (Bacteriol. Proc., p. 92, 1965) thatreduced, endogenous cytochromes were thephotosensitizers.To obtain a more precise action spectrum, the

Argonne National Laboratory biological spec-trograph was employed. The data (5) yielded anaction spectrum clearly resembling the generalizedabsorption spectrum of porphyrins (16).

If stationary-phase cells of M. xanthus containa photosensitizer not present in logarithmic-phase cells, it should be possible to photosensitizethe latter by adding this material. It was initiallyobserved that light-insensitive, exponentiallygrowing cells were, themselves, photolysed whenadded in the light to a photolysed culture ofstationary-phase cells grown in the dark. Fur-thermore, the sensitizing moiety was released byphysicial disruption of sensitive cells, as wasdemonstrated by addition of sonic-treated prepa-rations to logarithmic-phase cells. Photosensitivecells were sonic-treated and centrifuged at 10,000x g, and the supernatant fraction was centri-fuged at 100,000 X g. Most of the photosensitiz-ing material was localized in the 100,000 X gpellet, suggesting that it may be bound to thewall-membrane complex.An experiment was carried out to determine

whether stationary-phase cells grown in the light,as well as logarithmic-phase cells, could be photo-sensitized exogenously and whether photosensi-tizer was present in sonic-treated preparations oflogarithmic phase and light-grown stationary-phase cells, as well as in preparations from cellsgrown in the dark. The data (Table 2) indicatethat the stationary-phase cells grown in the lightare insensitive to exogenous, as well as endoge-nous, photosensitizer. That these cells do have aphotosensitizer was confirmed by the ability ofa sonic-treated cell preparation to photosensitizelogarithmic-phase cells. A detectable amount ofphotosensitizer was not released when logarith-mic-phase cells were disrupted.Chemical reduction of 100,000 X g pellet

preparations of photosensitive cells with Na2S204did not significantly enhance their photosensitiz-ing activity, nor did oxidation with air or KsFe-(CN)6 destroy the activity. Also, ammoniumsulfate fractions of 100,000 X g supernatant andpellet preparations demonstrated no relationshipbetween the ability to photosensitize and theheight of 425-m,i peaks in difference spectra.

538 J. BAcTERioL.

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

LYSIS AND CAROTENOGENESIS IN M. XANTHUS

TABLE 2. Photosensitizationi of Myxococcus xan-thus FBt logarithmic-phase cells by particulatefractions of sonic-treated preparations of

dark- and light-grown stationary-phasecells

Cells Sonic-treated prepn lighti

Logarithmic .No lysisStationary, dark LysisStationary, light-. No lysisLogarithmic ........ Logarithmic No lysis

cellsLogarithmic......... Stationary, Lysis

darkLogarithmic. Stationary, Lysis

lightStationary, light.... Logarithmic No lysis

cellsStationary, light Stationary, No lysis

darkStationary, light .Stationary, No lysis

light

* Incubation in the dark consistently producedno lysis.

Both experiments suggested that the photosensi-tizing moiety was not a reduced cytochrome(s).An experiment was carried out in which FBt

cells grown in the dark were sampled at varioustimes during the stationary phase. Each samplewas adjusted to contain essentially equal numbersof cells and was assayed for: rate of endogenousphotolysis (Fig. 3), rate of exogenous photo-sensitization of logarithmic-phase cells by 10,000X g supernatant fractions of sonic-treated cells(Fig. 4), difference spectrum of 10,000 X g super-natant fraction (Fig. 5), and spectrum of 10,000 xg supernatant fraction (Fig. 6). A 500-ml cultureof FBt was sampled seven times from onset ofthe stationary phase (zero-time) to 31 hr intothe stationary phase.

Figure 3 demonstrates increasing rates ofendogenously sensitized photolysis with ageduring the stationary phase.

Figure 4 provides evidence that, as the endoge-nous photosensitivity of the cells increased, theability of cell-free preparations exogenously tophotosensitize insensitive cells also increased.The 425-m,u peaks of difference spectra of the

10,000 X g supernatant fractions are depictedin Fig. 5. There is little detectable difference inheight of the peaks, except at 31.3 hr, indicatingno correlation between absorption at 425 m,u andincreasing photosensitivity (Table 3).

Figure 6 demonstrates that the amount ofpigment absorbing at 410 m,u increased markedlyduring the stationary phase. Lesser peaks appear

109

0

0

O HlBODW-29

.5 HRSOD*e.34

O 48 96 144TIME OF ILLURMNATION (MINUTES)

FIG. 4. Photolysis of logarithmic-phase cells bypreparations from dark-grown stationary-phase cells ofincreasing age.

z

5 HRS

19.3 HRS

6314RS6 HRS3.5HRS1.5HRS

0 HRS

375 400 425 450 475 500WAVELENGTH (MO)

FIG. 5. Difference spectra of 10,000 X g superna-tant fractions ofdark-grown Myxococcus xanthus FBtcells sampled at different times after onset of thestationary phase.

at 507, 542, 579, and 633 m,. There is a relativelyconstant ratio of the height of the 410-mi, peakto the slope of the first phase of decline of loga-rithmic-phase cells exogenously photosensitized

VOL. 91) 1966 539

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

BURCHARD AND DWORKIN

04

0.2

0

40 <40 500 550 00 650

HVB (M M)

FIG. 6. Spectra of 10,000 X g supernatant fractionsof dark-grown Myxococcus xanthus FBt cells sampledat different times after onset of the stationary phase.

TABLE 3. Correlation ofspectral peaks of10,000 Xg supernatant fractions of dark-grown station-

ary-phase Myxococcus xanthus FBt cellswith rates of exogenous photosensiti-

zation of logarithmic-phase FBtcells by these fractions

Optical densityI

Sample Slp ffrtColumn ColumnStimmle hase of log At 425 m,., At 410tie declime educe At41

(Fig. 4) reu d MIoxdized (Fig. 6)

hr

0 0.59 0.035 0.29 0.059 0.491.5 0.42 0.035 0.28 0.083 0.673.5 0.70 0.025 0.34 0.036 0.495.8 0.65 0.03 0.42 0.046 0.658.3 1.06 0.025 0.61 0.024 0.5819.3 1.65 0.03 0.72 0.018 0.4431.3 2.90 0.09 0.98 0.031 0.34

by 10,000 X g supernatant preparations (Table 3).Variation in the value of the ratio may reflectthe manner in which the rate of photosensitiza-tion was calculated.The spectrum of these preparations and the

action spectrum (5) are similar to the absorptionspectrum of porphyrins. Therefore, an attemptwas made to isolate porphyrin(s) from photo-sensitive cells. The 1.0 N HCI extract containeda pigment with an absorption spectrum depictedin Fig. 7.

04 4

02 oa001 ,I

36 40 4 46 Wo 0 m 60 06 660WAVE LENGTH (MYJ)

FIG. 7. Spectrum of protoporphyrin IX from sta-tionary-phase Myxococcus xanthus FBt cells (solvent:I N HCI).

The 408-m,.t Soret band and smaller peaks at556 and 602 m,u were similar to the spectrum ofprotoporphyrin IX. Spectroscopy of a dioxanesolution demonstrated peaks at 405, 503, 537,575, and 630 m,, which were identical with thoseof protoporphyrin IX. Additional techniques (24)were used to confirm the identification. The pig-ment fluoresced red when excited with ultravioletlight. Upon fractionation with butanol-10%NaOH, the butanol phase fluoresced, indicatingthat the porphyrin contains two carboxyl groups.The pigment was not extractable from ethylacetate with 1% HCl, but it was extracted with5% HCl, giving an approximate HCI numberconsistent with the properties of protoporphyrinIx.Equal quantities of light-insensitive logarith-

mic-phase cells and light-sensitive stationary-phase cells of FBt were extracted for proto-porphyrin IX. A small amount of the pigmentwas found in logarithmic-phase cells; approxi-mately 16 times this amount was found in sta-tionary-phase cells. To determine whether theextracted protoporphyrin IX could photosensi-tize logarithmic-phase cells, 0.2 ml of a solutioncontaining approximately 1 ,g/ml and broughtto pH 7.8 with NH40H was added to 0.3 ml oflogarithmic-phase FBt cells in CT-1 at an ap-proximate concentration of 4 X 108 cells permilliliter. The data are presented in Table 4.The isolated protoporphyrin IX photosensitizeslogarithmic phase cells, and an authentic crystal-line sample of protoporphyrin IX also acted as aphotosensitizer.

Spectroscopy of the isolated protoporphyrinIX at pH 7.8 demonstrated no characteristicpeaks, nor did the solution fluoresce. Reacidifica-

TABLE 4. Photosensitization of logarithmic,phaseMyxococcus xanthus FB, cells by protoporphyrinIX extracted from stationary-phase FBi cells

Protoporphyrin IX Incubation Result

+ Light Lysis+ Dark No lysi$_ Light No lysis

Dark No lysis

540 JL BACSEREOL.

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

LYSIS AND CAROTENOGENESIS IN M. XANTHUS

tion resulted in the return of the characteristicbehavior.

Pigment characterization. The predominant pig-ment in acetone extracts of stationary-phaseFBt cells grown in the light was a carotenoid withabsorption maxima at 455, 477, and 510 m,u.FBy extracts had additional peaks at 360, 373,and 405 m,u; FBt extracts demonstrated ab-sorption in this region, but distinct peaks wererarely detectable.

Extracts of stationary-phase FBt cells grownin the dark demonstrated absorption maxima at361, 379, 403, 504, 537, 579, and 633 m,u. FByextracts had almost identical maxima. However,the relative heights of the peaks varied. The yellowcolor of stationary-phase FBy cells grown in thedark was probably due to its high content ofpigment absorbing at 379 m,u.

Appearance of pigment. Pigmentation, as wellas photosensitivity, also appeared after termina-tion of the logarithmic phase. Figure 8 shows thecontents per cell of pigments absorbing at 379and 477 m,i after different lengths of incubation(acetone extracts of wet cell pellets). The pointat which logarithmic growth terminated is indi-cated. No significant amount of either pigmentwas present until the onset of the stationaryphase, when a sharp rise in pigment synthesisoccurred. There was no detectable carotenoidsynthesis at 477 m,u in either strain when grownin the dark. Other experiments demonstratedthat a pigment absorbing at 403 m,u, also appear-ing during the stationary phase, developed to thesame level per cell in light and dark cultures.

Fruiting-body pigmentation. Myxococci areoften identified by the color of their fruitingbodies. It was of interest to examine the locationof pigments in these multicellular structures.Greene and Leadbetter (Bacteriol. Proc., p. 34,1962) reported that M. xanthus microcysts con-tain smaller amounts of saturated carotenoidsand larger amounts of less-saturated (colored)carotenoids than do vegetative cells. Strain FBy,because of its higher pigment content, was usedfor fruiting-body experiments, as described byDworkin (8). The pigmentation of fruiting bodiesincubated in light or dark was qualitatively thesame as that found in the respective stationary-phase vegetative cells grown in light or dark.Sonic treatment of a suspension of these fruitingbodies, containing vegetative cells as well asmicrocysts, released the pigment into the 5,900 Xg supernatant fraction; unpigmented microcystsresistant to sonic treatment were separated fromthe suspension by centrifugation. Therefore, es-sentially all of the 477-m,u pigment in light-incubated, and 379-m,u pigment in dark-incubated,

1.5r FBT

25 4 35 45TIME OF INCUBATION (HOURS)

FIG. 8. Development of pigment in Myxococcusxanthus FBt and FB, during light and dark growth.Symbols: 0, dark-grown, 379 m,u; A, dark-grown,477 m,u; Q, light-grown, 379 m,u; A, light-grown, 477mI.

6- to 7-day-old fruiting bodies is associated withvegetative cells.

Photoinduction of carotenogenesis. Incubationof cells in various combinations of light and darkperiods indicated that illumination was not re-quired throughout the period of carotenogenesis.Furthermore, photoinduction of carotenogenesisappeared to be independent of light intensity overa range of 11 to 165 ft-c of light passing throughyellow gelatin filters. This suggested that photo-induction sites were saturated at a low lightintensity.A crude action spectrum for carotenogenesis

was carried out with filters described by Burchard(Ph.D. Thesis, Univ. Minnesota, 1965). In de-termining the radiant power actually reachingthe cells at any one wavelength or in any band,both the spectrum of emission of the lamp and thebroad spectrum of transmission of the respectivefilters had to be considered. The data suggestedthat light in a band between 390 and 425 m,uand in the green region of the visible spectrumstimulated carotenogenesis.

Carotenoid function. It was of interest to deter-mine whether the function of the carotenoid inM. xanthus is a photoprotective one as it is inother microorganisms containing these pigments.

541VOL. 91, 1966

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

BURCHARD AND DWORKIN

TABLE 5. Correlation ofappearance ofcarotenoid inlight-grown cultures ofMyxococcus xanthus FBt

with development ofphotosensitivityin dark-grown cultures

Sample Carotenoid (477-mM Rate of photolysis Rate of lysistSmPle absorption) in of dark-grown per carotenoidtime light-grown cells cells content

hr

0 0.136 0.32 2.42.5 0.261 0.54 2.14 0.355 0.70 2.06 0.496 1.45 2.98 0.595 1.43 2.410.8 0.732 1.90 2.6

It was indicated above that carotenoid absorbingat 477 m,u appeared in light-grown cells im-mediately prior to the time of appearance ofphotosensitivity in dark-grown cells. To deter-mine whether there was a correlation betweenappearance of photosensitivity in dark-growncells and appearance of carotenoid in light-growncells, two FBt cultures were inoculated withequivalent amounts of exponentially growingFBt cells and incubated in light and dark. Atintervals, samples were withdrawn from the light-grown culture for assay of carotenoid content(477-m,u absorption) and from the dark-grownculture for determination of the rate of photoly-sis, calculated as the slope of the curve during thefirst exponential phase of decline (A log of totalcell number/A time). The data are presented inTable 5. Any variation from the relativelyconstant ratio of rate of photolysis of dark-grown cells to carotenoid content of light-grown cells could not be correlated with age.

Further evidence regarding the photoprotec-tive role of carotenoids arose from experimentsin which diphenylamine was used to inhibitsynthesis of unsaturated carotenoids. Dark-grown stationary-phase FBt cells were light-sensitive with or without diphenylamine in thegrowth medium. Cells grown in white and redlight in the absence of diphenylamine developed477-mis carotenoid and were found to be in-sensitive to light. Cells grown in red light in thepresence of 10-4M diphenylamine developed only1O0% of the carotenoid found in cells grown inwhite light without diphenylamine; the formerwere photolysed by exposure to white light.FBt dc, the mutant which synthesizes carote-

noid during the stationary phase in both light anddark, was insensitive to light unless carotenoidsynthesis was inhibited by growth in diphenyl-amine.Dworkin (7) and Mathews (20) demonstrated

a loss of photo-protection when carotenoid-

cr108 0\

4 DARK GROW,;FZ IV LIH- 11220ROM.~~~~~~~~~~~~40

MMR GROW*.40

106 II0 25 50 75 100 125 ISO 175

TIME IN MINUTES

FIG. 9. Lysis of light- and dark-grown stationary-phase Myxococcus xanthus FBt cells by high intensitylight at 4 C. Solid line represents dark controls at 4and 22 C.

containing Rhodopseudomonas spheroides and aMycobacterium sp. were illuminated in the cold.A similar experiment was carried out with sta-tionary-phase FBt cells grown in the light (Fig.9). It was found, by use of a high-intensity lightsource (approximately 18,000 ft-c), that ca-rotenoid-containing cells photolysed relativelyrapidly at 4 C. Dark-grown cells photolysed ex-tremely rapidly at the same temperature. Pho-tolysis of the light-grown cells also occurred at22 C, albeit at a considerably lower rate than at4 C.

DiscussIoNThe photolysis of dark-grown stationary-phase

cells of M. xanthus is a typical photodynamicprocess. It is purely photochemical, involving nodiffusional or enzymatic steps, as demonstratedby adherence of the process to the reciprocitylaw and by its temperature independence.

Burchard and Dworkin (Bacteriol. Proc., p.92, 1965) suggested that endogenous reducedcytochromes are the photosensitizers of station-ary-phase M. xanthus cells. The suggestion wasbased on a crude photolysis action spectrum,which resembled the difference spectrum ofcytochromes of M. xanthus (9), and on-the loca-tion of the photosensitizer in a particulate frac-tion of the cells. Further experiments attemptingto increase and decrease photosensitization in100,000 X g pellet preparations of photosensitive

542 J. BACI-E IOL.

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

LYSIS AND CAROTENOGENESIS IN M. XANTHUS

cells by chemical reduction and oxidation, re-spectively, were inconclusive. If reduced cyto-chromes were the photosensitizing moieties, adecrease in the height of the 425-mM Soret peakof difference spectra would be expected with in-creasing age and photosensitivity. The data inTable 3 do not support this suggestion. However,there is the possibility that in the preparation ofthe fractions cytochromes are oxidized to thesame level by 02A refined photolysis action spectrum, isola-

tion of protoporphyrin IX from photosensitivecells, and photosensitization of light-insensitivecells by protoporphyrin IX have since demon-strated that protoporphyrin IX is the photosensi-tizing moiety in stationary-phase M. xanthuscells. Furthermore, the photoresistance of loga-rithmic-phase cells is correlated with the pres-ence of relatively small quantities of protopor-phyrin IX.Blum (2) suggested that endogenous proto-

porphyrin photosensitizes red blood cells. Rud-zinska and Granick (23) found that old culturesof Tetrahymena geleii produce protoporphyrinafter growth in the dark. Upon exposure to light,cells containing the pigment either lysed or thepigment was released into the medium. Granick(12) identified, as protoporphyrin IX, a pigmentwhich accumulates in an X-ray induced mutantof Chlorella after the termination of logarithmicphase. Lascelles (15) discussed microorganismswhich form considerable amounts of free por-phyrin when grown under conditions limitingsynthesis of heme enzymes. Thus, protoporphyrinhas been implicated as an endogenous photo-sensitizer; furthermore, in some organisms freeporphyrins appear in quantity only durin-, thestationary phase or at times of biosyntheticlimitation. We suggest that protoporphyrin IXin M. xanthus may be a precursor or breakdownproduct of particle-bound cytochromes, and itsappearance may reflect a biosynthetic alterationresulting from respiratory changes occurring inthe stationary phase.Any molecule lacking significant light absorp-

tion and unable to fluoresce would not be expectedto photosensitize (2). The loss of fluorescence andtypical absorption spectrum of isolated proto-porphyrin IX in the physiological pH range,probably due to colloidal aggregation, suggeststhat association with the wall-membrane com-plex of the cell may be necessary for the char-acteristic absorption and fluorescence properties.An alteration of the absorption spectrum ofchlorophyll upon isolation has been reported byWassink et al. (27). To account for the observedexogenous photosensitization by protoporphyrinIX, it is suggested that absorption of the mole-

cules to the wall-membrane complex of viablecells simulates its in vivo state in stationary-phasecells, resulting in the reappearance of the char-acteristic absorption spectrum, its ability tofluoresce, and its photosensitizing ability.Our results indicate that the site of photo-

damage is the cell wall-membrane complex. Thefacts, that 100,000 X g pellet fractions of sonic-treated photosensitive cells can photolyse insen-sitive cells and that morphological changes oc-curring during cell photolysis indicate alterationof the limiting membranes of the cells, suggestthat the photosensitizer is bound to the wall-membrane complex. Also, the carotenoid must beclosely associated with the photosensitizer toserve in a photoprotective role; the pigment isapparently located in the wall-membrane complex(18).

It appears that divalent cations maintain awall-membrane macromolecular configurationsimilar to that present in a medium of low ionicstrength, possibly by forming salt bridges be-tween amino acid carboxyl groups as was sug-gested by Brown (4) for a marine pseudomonad.This configuration may be such that immediatecontact of structural components with boundphotosensitizer is prevented. Monovalent cationsin the absence of divalent cations may alter theconfiguration of membrane-bound structuralcomponents, placing them in apposition tophotosensitizing moiety and subjecting them tophotooxidation.

Further study is required regarding two ob-servations. (i) Visible photodamage appears tobe divisible into light-dependent and monovalentcation-dependent phases. It may be that light,in the absence of monovalent cation, may causedamage which is not visibly manifested. (ii)Cells grown in CT, with twice the MgSO4 con-centration as in CT-1, are not photolysed. Thiswould suggest an alteration in the wall-membranecomplex. McQuillan (17) has reviewed severalstudies which indicate that Mg++ functions inmaintaining the integrity of protoplast membranesand preventing penicillin lysis in gram-positiveand gram-negative microorganisms. The fact thatcells do photolyse in CT-1, containing 1.85 X10-3 M MgSO4, suggests that the photoprotectiveMg++ may be chelated by medium constituents.The spectra of acetone-soluble pigments of M.

xanthus FBt and FB, correlates well with that ofMyxococcus pigments described by Greene andLeadbetter (Bacteriol. Proc., p. 34, 1962; p. 54,1963; p. 17, 1964). The orange carotenoid in light-grown cells appears to be Greene and Lead-better's P-476 or P-478. The predominant pig-ment(s) of dark-grown cells (maxima at 361 and379 m,u) may be identical to Greene and Lead-

543VOL. 9 1 1966

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

BURCHARD AND DWORKIN

better's P-355 and P-378. These authors (Bac-teriol. Proc., p. 17, 1964) suggested that P-378is a precursor for the light-induced P-478. The403-m, peak in acetone extracts of wet pellets ofstationary-phase cells is identical to the 410-m,upeak predominant in aqueous suspensions ofcell-free preparations; we suggest that it is theSoret band of protoporphyrin IX.As in other carotenoid-containing microor-

ganisms the pigment(s) in M. xanthus are photo-induced, but only at a time immediately prior tothe appearance of photosensitivity in dark-growncells. The increase in carotenoid content in thelight parallels increased photosensitivity in thedark, suggesting a finely controlled regulation ofthe amount of photoprotective pigment.From a crude action spectrum we would sug-

gest that the photoreceptor for carotenogenesiscould also be protoporphyrin IX. A sublethalamount of the porphyrin is present in light-insensitive cells; relatively little light in contrastto that required for photolysis is required forcarotenogenesis. The initial photoinduction isprobably followed by light-independent darkreactions, as suggested for Mycobacterium caro-tenogenesis (22).

It has been proposed that carotenoids functionas a substrate for photooxidation by the excitedphotosensitizer in various microorganisms; theymay then be cycled back to the reduced form byan enzymatic mechanism (1, 7, 10, 20, 26). Evi-dence for the same function of carotenoid in M.xanthus has been presented in the form of anexperiment which demonstrates photo-sensitiza-tion of carotenoid-containing cells at high lightintensity at 4 C. It is suggested that the unsatu-rated carotenoid absorbing at 477 m,u is inter-posed between the photosensitizer and the photo-oxidizable structural component of the cellwall-membrane complex; the carotenoid may thenserve as an alternative substrate for photooxida-tion. A reductase serves to cycle the oxidizedcarotenoid to the reduced state. This activity isinhibited at low temperature. Upon exposure tolight, carotenoid-containing cells incubated at 4C or dark-grown stationary-phase cells depletetheir oxidizable pigment or have none, respec-tively; they then photolyse.

Exposure to light may have provided selectivepressure for the development of carotenoids inM. xanthus and their role in photoprotection.However, presence of photosensitizer is onlymanifested in stationary-phase cells. In theecological niche of Myxococcus, is it likely thatthe organism ever enters the stationary phase?In batch cultures, the stationary phase of growthis usually a consequence of the depletion of anessential component for growth or of the ac-

cumulation of toxic end products. The abilityof the organism to move over a solid surface inits natural environment and its ability to formfruiting bodies and microcysts under conditionsof nutrient depletion suggest that it need neverencounter either of these situations in the vege-tative cell state. When, then, does photosensi-tivity appear in nature? Also, the question arisesas to whether the organism encounters the ionicconditions required for manifestation of photoly-sis in its environment. The ecological significanceof this photosensitivity is, therefore, not clear.Yet, assuming that the primary role ofcarotenoidsis a photoprotective one, the organism's abilityto synthesize carotenoid(s) would indicate thatit encounters conditions in which it is potentiallyphotosensitizable.

ACKNOWLEDGMENTS

This investigation was supported by grant GB-9from the Developmental Biology Program of the Na-tional Science Foundation and by Public HealthService training grant AI-90-05 from the NationalInstitute of Allergy and Infectious Diseases.

LITERATURE CITED

1. BAMJI, M. S., AND N. I. KRINSKY. 1965. Caro-tenoid de-epoxidations in algae. II. Enzymaticconversion of antheraxanthin to zeaxanthin.J. Biol. Chem. 240:467-470.

2. BLUM, H. F. 1941. Photodynamic action and dis-eases caused by light. Reinhold PublishingCorp., New York.

3. BOGORAD, L., AND S. GRANICK. 1953. Protopor-phyrin precursors produced by a Chlorellamutant. J. Biol. Chem. 202:793-800.

4. BROWN, A. D. 1963. The peripheral structures ofgram-negative bacteria. IV. The cation-sensi-tive dissolution of the cell membrane of thehalophilic bacterium, Halobacterium halobium.Biochim. Biophys. Acta 75:425-435.

5. BURCHARD, R. P., S. A. GORDON, ANm M. DWoR-KIN. 1966. Action spectrum for the photolysisof Myxococcus xanthus. J. Bacteriol. 91:896-897.

6. DUNDAS, I. D., AND H. LAREN. 1963. A study onthe killing by light of photosensitized cells ofHalobacterium salinarium. Arch. Mikrobiol.46:19-28.

7. DWORKIN, M. 1959. Function of' oids nphotosynthetic bacteria. Nature 184:1891

8. DWORKIN, M. 1962. Nutrtional r entsfor vegetative growth of Myxococcus xanthus.J. Bacteriol. 84:250-257.

9. DWORKIN, M., AND D. J. NEDERPRUEM. 1964.Electron transport system in vegetative cellsand microcysts of Myxococcus xanthus. J.Bacteriol. 87:316-322.

10. FELDMAN, L. A., AND E. S. LmINDSomR. 1964. Theeffect of carotenoid pigments on photooxida-

544 J. BA<RI.

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

LYSIS AND CAROTENOGENESIS IN M. XANTHUS

tions of some photosynthetic bacteria. Bio-chim. Biophys. Acta 79:266-272.

1 1. GOODWIN, T. W., AND H. G. OSMAN. 1953. Studiesin carotenogenesis. 9. General cultural condi-tions controlling carotenoid (spirilloxanthin)synthesis in the photosynthetic bacterium Rho-dospirillum rubrum. Biochem. J. 53:541-546.

12. GRANICK, S. 1948. Protoporphyrin as a precursorof chlorophyll. J. Biol. Chem. 172:717-727.

13. GRIFFITHS, M., W. R. S1STROM, G. COHEN-BA-ZIRE, AND R. Y. STANIER. 1955. Function ofcarotenoids in photosynthesis. Nature 176:1211-1214.

14. KUNISAWA, R., AND R. Y. STANER. 1958. Studieson the role of carotenoid pigments in a chemo-heterotrophic bacterium, Corynebacterium poin-settiae. Arch. Mikrobiol. 31:146-156.

15. LASCELLES, J. 1955. The formation of porphyrinsby photosynthetic bacteria. Ciba Found. Symp.Porphyrin Biosyn. Metab., p. 265-284.

16. LASCELLES, J. 1964. Tetrapyrrole biosynthesis andits regulation. W. A. Benjamin, Inc., New York.

17. MCQUILLEN, K. 1960. Bacterial protoplasts, p.249-359. In I. C. Gunsalus and R. Y. Stanier[ed.], The bacteria, vol. 1. Academic Press, Inc.,New York.

18. MASON, D. J., AND D. POWELSON. 1958. The cellwall of Myxococcus xanzthus. Biochim. Biophys.Acta 29:1-7.

19. MATHEWS, M. M. 1963. Studies on localization,function, and formation of the carotenoid pig-

ments of a strain of Mycobacterium marinum.Photochem. Photobiol. 2:1-8.

20. MATHEWS, M. M. 1964. The effect of low temper-ature on the protection by carotenoids againstphotosensitization in Sarcina lutea. Photo-chem. Photobiol. 3:75-77.

21. MATHEWS, M. M., AND W. R. SISTROM. 1960. Thefunction of the carotenoid pigments of Sarcinialutea. Arch. Mikrobiol. 35:139-146.

22. RILLING, H. C. 1962. Photoinduction ofcarotenoidsynthesis of a Mycobacterium sp. Biochim. Bio-phys. Acta 60:548-556.

23. RUDZINSKA, M. A., AND S. GRANICK. 1953. Proto-porphyrin production of Tetrahymena geleii.Proc. Soc. Exptl. Biol. Med. 83:525-526.

24. SCHWARTZ, S., M. H. BERG, I. BOSSENMAIER, ANDH. DINSMORE. 1960. Determination of porphy-rins in biological materials. Methods Biochem.Anal. 8:221-293.

25. SISTROM, W. R., M. GRIFFITHS, AND R. Y. STA-NIER. 1956. The biology of a photosyntheticbacterium which lacks colored carotenoids. J.Cellular Comp. Physiol. 48:473-515.

26. STANIER, R. Y., AND G. COHEN-BAZIRE. 1957. Therole of light in the microbial world: some factsand speculations. Symp. Soc. Gen. Microbiol.7:56-89.

27. WASSINK, E. C., E. KATZ, AND R. DORRESTEIN.1939. Infrared absorption spectra of variousstrains of purple bacteria. Enzymologia 7:113-129.

VOL. 91, 1966 545

on April 9, 2018 by guest

http://jb.asm.org/

Dow

nloaded from