JOURNAL OFCLINICAL MICROBIOLOGY, Nov. 1987, p. 2054-2058 Vol.25, No. 11 0095-1137/87/112054-05$02.00/0
Copyright C 1987, American Society forMicrobiology
Colony
Formation and
Morphology
in
Borre
lia
burgdorferit
TIMOTHY J. KURTTI,' ULRIKEG.
MUNDERLOH,1*
RUSSELL C.JOHNSON,2
AND GILBERT G.AHLSTRAND3
Departments ofEntomology' and PlantPathology,3 University ofMinnesota, St. Paul, Minnesota55108,
and Departmentof
Microbiology,
Universityof
Minnesota,Minneapolis,
Minnesota 554552Received 22 June 1987/Accepted 28 July 1987
Twostrains of Borreliaburgdorferi, B31 and 297,formed colonies whenplatedontoBarbour-Stoenner-Keiy mediumsolidified with agarose(1.3%) and incubatedin acandlejarat34°C.Coloniesdifferinginmorphology were observedinbothstrainsafter2 to3weeksof incubation. Strain B31 colonieswereeithercompact,round (meandiameter, 0.43 mm),andrestrictedtothesurface oftheagarose mediumordiffuse(mean
diameter,
1.80 mm)and penetrating intothe solid medium. Strain 297colonies meani diameter, 1.43mm)either showed a raised centersurroundedbyadiffuse ringofspirochetesorconsistedofnumeroussmallspirochetal
aggregates. Bothcolony typesexpanded intotheagarosemedium.Scanningelectronandlightmicroscopyconfirmed that thecolonieswereformedby spirochetes. Twistedtanglesof intertwinedspirocheteswere visibleonthesurface, withnumerousspherical bodies amongthem,especiallyin the centralregions.Attheperiphery,the borreliae were moreloosely packed, and individual coilswerediscernible.Borrelia burgdorferi, the etiological agent of Lyme
dis-ease, infectsanumberof vertebrates, includinghumans and
domestic as well as feral animals (1, 2). The
spirochetes,
transmitted primarilyby ticks ofthe genusIxodes(7, 9, 10),
were first isolated from the midgut ofa deer tick, Ixodes
dammini, in modified Kelly medium (8). Variations ofthis
liquid Barbour-Stoenner-Kelly (BSK)mediumareroutinely
used toculturethespirochetes in vitro (3, 4). When supple-mented with certain antibiotics, it permits the selective
isolation of spirochetes from tick tissues (14). Strains that
have been extensively passaged in vitro can be grown in
liquid BSK medium from single spirochetes,i.e., cloned(4),
but thegrowth ofspirochetes asisolatedcolonieson asolid
medium has not been achieved. Barbour (3) described the growth of B. burgdorferi as a "lawn" on BSK medium
solidified with agarose, but colony formation was not
re-ported. For theselection ofvariants, spontaneous mutants,
and recombinant clones carrying specific DNA inserts, an
agar-cloning procedureisanecessity.Inthis communication
wepresent aculture system thatpermits colonial growthof
B. burgdorferi.Thegrowthofthecoloniesand theirlightand
electronmicroscopic appearance aredescribed.
MATERIALS AND METHODS
Spirochetes. Two strains of B. burgdorferi were used: strain B31, isolated from I. dammini (8) and previously clonedby limiting dilution(4), and the uncloned 297 strain,
isolated from human spinal fluid(18).
Culture media. Spirochetes were routinely maintained in
liquid BSK medium prepared as described by Barbour (3).
Cultureswereincubated at 34°C and transferred at intervals
of 7to 10 dayswitha 1% (vol/vol) inoculum.
The solid medium was prepared as follows. Fraction 1: to 900 ml ofwater(18
MfQ
resistivity Milli-Q water; MilliporeCorp., Bedford, Mass.) were added 50 g
of
bovine serumalbumin
(fraction
V; Armour Pharmaceuticals, Kankakee,111.),
5 gof Neopeptone(DifcoLaboratories, Detroit, Mich.),6 g of HEPES
(N-2-hydroxyethylpiperazine-N'-2-ethane-*
Corresponding
author.tPaperno. 15,454, Scientific JournalSeries, Minnesota Agricul-turalExperiment Station.
sulfonic acid) (Sigma Chemical Co., St. Louis, Mo.), 0.7 g of sodiumcitrate, 5 g ofglucose, 0.8 g of sodium pyruvate, 0.4
gofN-acetylglucosamine (Sigma),2.2 g of sodium
bicarbon-ate, and 2.53 g of TC Yeastolate (Difco). This was filtered through0.22-,um-pore-size membranes and stored at4°C in 90-ml portions until needed. Fraction 2 (made fresh every time) 2.8 g ofgelatin (Difco)wasdissolved in 20 ml of water (60°C; Millipore), and then 1.7 g of agarose (SeaKem LE, lowelectroendosmosis; Polysciences, Inc., Warrington, Pa.) was added. The mixture was solubilized and sterilized by autoclaving (15 minat121°C) and thenkeptwarmin a60°C water bath. Fraction 1 and heat-inactivated rabbit serum (Pel-FreezBiologicals, Rogers, Ariz.)were also warmed to 60°C in thewaterbath.All solutionswerethen transferred to a heating plate inside a laminar flow hood, where the final mediumwas madeby mixing90 ml offraction 1, 6.4 ml of rabbit serum, and 20 ml of the gelatin-agarose solution. Thorough mixing of the warm medium components was necessary toensuretheircompleteblending. Finally, 10 ml ofprewarmed CMRL (ConnaughtMedical Research
Labo-ratories) 1066 without glutamine (GIBCO Laboratories,
Grand Island, N.Y.)and 5.3 ml ofa 5% aqueous NaHCO3 solutionwere added. Thismediumwasdispensed into petri dishes (Nunc, Roskilde, Denmark; 35-mm diameter; 2to 3 ml perplate),and theplateswereallowed tocoolanddry for
15min ina laminar flowhood.
Inoculation and incubation of plates. Spirochetes from liquid cultures were counted in a Petroff-Hausser bacteria
countingchamber. Culturescontaining 107 to 108 spirochetes
per ml were serially
diluted
10-fold to give a range of dilutions fromabout1,000 to less than 1spirochete per 25 ,il. For each dilution, threereplicate plates were inoculated by placing25,ul ofspirochete suspension onto the surface of the agarose medium. The dishes weretilted to spread the fluidevenlyontothesurface and then placed in a glass desiccator
jarwithastopcockonthe lid. Each jar contained 50 to 100 ml of distilled water on the bottom and a white paraffin candleontheporcelain plate. The candle was lit and the jar wascovered, leaving the stopcock open. Immediately after theflamesubsided,thestopcock was closed and the jar was
placedina34°C incubator. After 2 to 4 weeks Qf incubation,
the plateswere examined for the presence of colonies by a
dissecting microscope with transmitted light. The colonies
2054
on April 11, 2020 by guest
http://jcm.asm.org/
werecounted, and their numbers werecomparedwiththose
obtained by direct enumeration inthePetroff-Hausser
cham-ber.
Stabilityof colonymorphology. To obtain an indication of
the phenotypic stability of colony morphology, well-isolated
colonies from each strain were selected by morphological
characteristics. Theywêre pickedwith aPasteur pipette and
transferred toliquidBSK. After 1 week of incubation, serial
10-fold dilutionsofspirocheteswere madefrom each clone,
and 25-,ul portions of each dilution were inoculated onto
fresh plates of agarose-solidified BSK and incubated in a
candlejaras outlined above.
SEM.Agaroseblocks containingindividualcolonies were
excised and fixed in 2.5% glutaraldehyde in 0.05 M
phos-phate
buffer (pH
7) at 4°C for 1 to 3 days. Samples werepostfixed
with the OTO(osmium-thiocarbohydrazide-osmium)
double
osmium tetroxide application method (15)and dehydrated inagraded ethanol series
of
20, 40, 60, 80,and 99%
ethaniol,
followed by three changes of absoluteethanolfor10 mnineach. Blocks were dried out of liquidC02
in a Ladd
critical-point
dryer. They were mounted onaluminum pin stubs with colloidal graphite paste prior to
coating ina
Kinney
vacuumevaporatorwithapproximately200 nm of60% gold-40% palladium metal. Samples were
viewed in a Philips
500X
scarining electron microscope(SEM) at 12 kV acceleration voltage, and the image was
recordedonPolaroid type
55 film.
RESULTS
Previous studies
(3)
in which B. burgdorferi failed toproducediscrete colonieson amedium solidified with0.8%
agarosesuggestedthat Lymediseasespirochetesmight need
afirmer mediumtorestrict their movements. Furthermore,
they
wereknown to require microaerophilic conditions forgrowth (5). To nieet these criteria, wedoubled the
concen-tration of gelatinandincreasedthe agaroseconcentration to
1.3%.
The candle jar was chosen because it is a simpledevicethat canprovideahumidified
atmosphere
of reduced02
and increasedC02
(13).
Colonieswerevisible after2weeksof incubation, but3 to
4weeks wereneeded foraccurateenumeration and
evalua-tion of morphology. With the unaided eye, the colonies
generally
appearedassmall, whitedisks,but whenexamined
with a dissecting microscope,
differences
in morphologybecame apparent.
On
plates
inoculated with B31 spirochetes, twotypes ofcoloniesdeveloped:
small,
compact, roundcolonieswith anaverage diameter of0.43 + 0.03 (standard deviation
[SDI)
mmwhichwererestrictedtothesurface(Fig. 1A),andlarger
diffuse colonies(1.80+ 0.57
[SD]
mm) whichpenetrated intotheagarose(Fig. 1B)to anaveragedepthof0.60 + 0.24(SD)
mm.The smallsurfacecolonieswerecomposed of tangles of
coiled
spirochetes
attheperiphery
andnumerousspherical
cells, along with verydensely packed coiled
spirochetes
inthe center
(Fig.
2). Theedges
ofsuchcolonies weresharp
and well
defined,
withnoisolated freespirochetes
presenton thesurrounding
agar surface. In contrast, diffuse coloniescontained fewer
spherical
bodies and were lesstightly
packed, and groupsof
spirochetes
were seenin the processof
migrating
awayfromthe
edges (not
shown).
Colonies formed
by
strain 297 were more variable inmorphology and intermediate forms were seen. Two main
types were discernible: diffuse colonies with or without a
conspicuous
raised center surroundedby
a diffuse ring ofspirochetes (Fig.
3A), and coloniescomprising
numerousFIG. 1. Twotypes ofB.
burgdorferi
B31 coloniesas seenundera
dissecting microscope
withtransmitted light
after 3 weeks of incubationat340C.
Bar,0.33mm.(A)Smaller,compactcolonywith distinct borders. (B) Larger colonywithdiffuse borders.spirochete
aggregates(granular
type)
(Fig.
3B).
Theraised-colony
centerscomprised
tightly
packed
andintertwined
spirochetes
(Fig.
4) sutrounded
by
afiat
layer
ofspirochetes.
After 3 weeks of
incubation,
the averagecolony
size was1.43 ± 0.71
(SD)
mm. All coloniespenetrated
into theagaroseto an average
depth
of0.57 ± 0.28(SD)
mm. Athigh
spirochete
concentrations(>500
per 25FjI),
the numberof colonies could notbeaccurately determined,
asthey
tendedtooverlap extensively
andform.lawns. Atlower concentrations ofspirochetes
(>100per25pil),
the number of colonies observed on aplate
correlatedapproximately
with the calculated number of
spirochetes
inoculated. The average number of colonies observed for eachspirochete
counted intheinoculum in thePetroff-Hausser chamberwas
1.12for strain' 297
(12
trials;
rangeofplating efficiency,
S to272%)
and 2.06for
strain'
B31 (7trials;
range ofplating
efficiency,
56to636%).
Well-isolatedcolonies subcultured into
liquid
medium andreplated
onto solid medium gave rise to colonies similar inmorphology
totheparental
type thatwaspicked.
The small,compact, round colonies of strain B31
again
formedsmall
colonies,
most of which were of the compact, round type,and othersweremore
diffuse,
whereasastrain B31colony
of thelarger
diffuse type gave riseagain
tolarge,
diffuse colonies. Strain 297 coloniesof eithertype
formed colonies ofbothmorphological
types afterrelating.
DISCUSSION
The
passage
ofborreliaethrough
vertebratehosts,
arthro-pod
vectors,
culture systemsand ispostulated
to causechanges
in theirphysiology,
antigenicity,
viability,
viru-lence,
andinfectivity (5).
Prolonged
invitrocultureMayleadto lossof
infectivity
forvertebrates,
while continuedpassage
from vertebrate to vertebrate host vianeedle inoculation of
infected blood reduces the
infectivity
for thearthropod
vector.
Antigenic changes
inthe
relapsing
feverspirochete
Rorrelia hermsii
within vertebrates occur at a ratehigher
than
canbe accounted
forby
mutationsandare achievedviatransposable
genes(6, 16,
17).
Whethersuchgenetic
mech-anisms are
alsoresponsible
forphernotypc
variability inother borreliae remains
to be determined. Tostudy
andcharacterize
thechanges
induced
by
environmentalcondi-tions
and
toseparate
anyvariants
from aheterogeneous
mixture,
theagar-cloni'ng
method has beenwidely
used.It
permits
one to linkmorphological
traits,
such ascolony
shape,
withphysiological
ones,
such asvirulence,
and toisolate
clones that
exhibit thedesired
characteristic.As
far
aswe
areaware,
this
isthe first
reportdescribing
colonial
growth
ofB.burgdorferi.
Barbour
(3)
cultured strainon April 11, 2020 by guest
http://jcm.asm.org/
2056 KURTTI ET AL.
w .
-FIG. 2. SEM of 3-week-oldB.burgdorferiB31colony, small,compact type,onagarose.Bar,30 ,um. Inset: Notethesharp,distinctborder (bar, 10 ,um).
B31on BSKmedium solidified with0.8% agarose. Cultures
incubated inacandlejargrewtoformalawnofspirochetes.
We used a higher
concentration
of agarose (1.3%) in BSK medium with twice theamount ofgelatin,but thecomposi-tionoftheatmospherewasthesame.Theseconditionswere
A
B
FIG. 3. B. burgdorferi 297 colonies as seen undera dissecting
microscope with transmitted light after 3 weeks of incubation at
34°C. (A) Colony with raisedcenteranddiffuse borders. (B) Colony comprisingnumerousaggregatesofspirochetes. Bar, 0.33mm.
suitable for the formation of discrete colonies by the
microaerophilic borreliae. This candle jar system, which
providesagaseousenvironmentconsistingof 3%C02, 17%
02,and80%N2 (13), is similartothat usedfor the cultivation
ofmalarialparasites in vitro. We also tested other
concen-trations ofagaroseandconfirmedthat0.8% isunsatisfactory
because of extensive migration ofthe spirochetes through
themedium. While 1.5%agaroseyielded
satisfactory
colonygrowth,itwas difficulttoevenly dissolve the agarosein the
BSK medium.
SEMobservationof wholecoloniesandlight microscopic
observation of subcultured
spirochetes
in fluid mediumconfirmed that the colonies were formed by spirochetes.
Also,the numberof colonies observed on platesinoculated
with dilutespirochetesuspensionsapproximated the number
ofspirochetes calculated to be present from direct count
data obtained with the Petroff-Hausser chamber. Various
methods have been usedtoestimate thenumberofborreliae
ina suspension(19). However, they do not evaluate
viabil-ity, nor can they be usedwith low numbers ofspirochetes.
The seemingly large variation inplating efficiency might be
explained by the fact that we used young, logarithmically
growing cultures with relatively low concentrations of
spi-J. CLIN.MICROBIOL.
on April 11, 2020 by guest
http://jcm.asm.org/
COLONIAL GROWTH OF BORRELIA BURGDORFERI 2057
t
vw
tX
rK
gu...<
gw~»~~~~~~~~~~~~~~~~~~...
rw*v
* X ir&t: « ' t «
('
<'"
FIG. 4. (A) SEM ofB. burgdorferi 297 colony with raised center and surrounding flat spirochete layer. (B) Masses ofintertwined spirochetesandspherical bodiesseeninthecenterofthecolony. Bars: 30p.m(A); 2,um (B).
VOL.25, 1987
R4
Ar,
ké,
on April 11, 2020 by guest
http://jcm.asm.org/
2058 KURTTI ET AL.
rochetes. Under such circumstances,countsobtainedbyuse
of thePetroff-Hausser chamber tendtobe lessaccurate.The method described here should overcome the difficulty of
enumeratingviable spirochetes and beapplicable togenetic
studiesaswellasantibiotic susceptibility assays.
We notedmorphological variations in the colonies formed between and within strains B31 and 297. The SEM observa-tions of selected colonies indicated that the spirochetes constituting the various colonies differed in their ability to spreadoverand migrate in the solid medium andtogrow as
aggregates. Thesecharacteristics may reflect important
dif-ferences inspirochete mobilityor adhesiveness thatcan, in
turn, have a bearing on their in vivo behavior, i.e., their tendencyto penetrate oradhereto the tissues of the verte-bratehost orvectortick. Spirochetes subcultured from the round,compactstrain B31 colonieswereindistinguishable in
liquid BSK medium from those transferred from a diffuse, spreading colony. Whether such differences in colony
mor-phologycanbe correlated with otherbiological features of B.
burgdorferi (e.g., virulence, infectivity, antigenicity) re-mains to be determined. Among leptospires, virulent and avirulent organisms aredistinguishable on the basis of col-ony morphology (11, 12). Diffuse colonies of Leptospira icterohaemorrhagiae werefound to contain leptospires vir-ulent for rodents, but small compactcolonies did not (12). Finally, by using clones derived from single agarose
colo-nies, wecanbe certain that variability inastrain isnot due to amixed population, but rather toan inherent genetic or
phenotypic variation.
ACKNOWLEDGMENTS
Thisresearchwas supported by the MinnesotaExperiment
Sta-tion andby Public HealthServicegrantAR-34744toR.C.J. from the NationalInstitutes ofHealth.
LITERATURE CITED
1. Anderson, J. F., R. C. Johnson, L. A. Magnarelli, and F. W.
Hyde. 1986. Involvement of birds in the epidemiology of the Lyme disease agent Borrelia burgdorferi. Infect. Immun. 51:394-396.
2. Anderson, J. F., and L. A. Magnarelli. 1984. Avianand
mam-malian hostsforspirochete-infected ticks andinsects inaLyme
disease focus inConnecticut. YaleJ.Biol. Med. 57:627-641. 3. Barbour, A. G. 1984.Isolation and cultivation ofLymedisease
spirochetes. Yale J. Biol. Med. 57:521-525.
4. Barbour,A.G., W.Burgdorfer, S. F. Hayes, O.Peter, andA.
Aeschlimann. 1983. Isolation ofa cultivable spirochete from Ixodesricinusticksof Switzerland. Curr. Microbiol. 8:123-126. 5. Barbour, A. G., and S. F. Hayes. 1986. Biology of Borrelia
species. Microbiol. Rev.50:381-400.
6. Barbour,A.G., andH.G. Stoenner. 1984.Antigenic variation of Borrelia hermsii. UCLA Symp. Mol. Cell. Biol. New Ser. 20:123-125.
7. Burgdorfer,W.1984.Discovery of the Lyme diseasespirochete and itsrelationtotickvectors. Yale J. Biol. Med. 57:515-520. 8. Burgdorfer, W.,A.G. Barbour, S. F. Hayes, J.L.Benach,E. Grunwaldt, andJ.P. Davis. 1982. Lyme disease-a tick-borne spirochetosis? Science 216:1317-1319.
9. Burgdorfer, W., and K. L. Gage. 1986. Susceptibility of the black-legged tick, Ixodesscapularis, to theLyme disease spi-rochete,Borrelia burgdorferi. Zentralbl. Bakteriol. Mikrobiol. Hyg. Abt.1Orig. Reihe B263:15-20.
10. Burgdorfer, W., R. S.Lane,A. G. Barbour,R. A.Gresbrink, andJ.R. Anderson.1985.Thewesternblack-leggedtick,Ixodes pacificus:a vectorof Borrelia burgdorferi. Am. J.Trop. Med. Hyg. 34:925-930.
11. Faine, S., and J. Van Der Hoeden. 1964. Virulence-linked colonial andmorphological variation in Leptospira. J. Bacteriol. 88:1493-1496.
12. Fujikura,T. 1966.Studieson twocolonial types of Leptospira icterohaemorrhagiaewithspecial referencetothebottle culture method. Jpn.J. Microbiol. 10:79-83.
13. Jensen, J. B.,and W.Trager.1977.Plasmodiumfalciparum in culture: use of outdated erythrocytes and description of the candlejar method.J. Parasitol.63:883-886.
14. Johnson,S. E., G.C. Klein, G.P. Schmid,G. S.Bowen, J. C. Feeley,and T. Schulze. 1984.Lymedisease:aselective medium for isolation of thesuspectedetiological agent,aspirochete. J. Clin. Microbiol. 19:81-82.
15. Kelley, R. O., R. A. F. Dekker, and J. G. Bluemink. 1973. Ligand-mediated osmium binding: its application in coating biological specimens for scanning electron microscopy. J. Ultrastruct. Res.45:254-258.
16. Meier,J. T.,M.I. Simon,and A.G. Barbour. 1985. Antigenic variation is associated with DNA rearrangements in arelapsing fever borrelia. Cell 41:403-409.
17. Plasterk, R. H. A., M. I. Simon, and A. G. Barbour. 1985. Transposition of structural genesto anexpression sequence on a linear plasmid causes antigenic variation in the bacterium Borreliahermsii. Nature(London) 318:257-263.
18. Steere, A. C., R. L. Grodzicki, A. N. Kornblatt, J. E. Craft, A. G. Barbour, W. Burgdorfer, G. Schmid, E. Johnson, and S. E. Malawista. 1983. The spirochetal etiology of Lyme dis-ease. N. Engl.J. Med. 308:733-740.
19. Stoenner, H. G. 1974. Biology of Borrelia hermsii in Kelly medium. Apple. Microbiol.28:540-543.
J.CLIN. MICROBIOL.
on April 11, 2020 by guest
http://jcm.asm.org/