Ac/a Tropica, 48(1991)89-94

Elsevier

ACTROP (K)llR4

Clinical implications of delayed growth of the Lyme borreliosis spirochete, Borrelia burgdor{eri

Alan R MacDonald!, Bernard W. Bcrger2, and Torn G. Schwan 3

"9

iDeparlmenz of Parholo,;)'. Sou/hampton Hosp/ial, Southampton, New York. NY. U.S.A .. 'Depanmelll o/Dermat%gy. New York University School 01 Medicine .. ""ew Y(}rk, ,I,'Y. 1/.S..4. and 'Department of Health and Human Services. Public Heaflh Senin, Nil/iona/lnstitute.I' "I Health. /''-arional/mlilUle of

Allergy and Infe('/ious Di,>ease.\, Arthropod-horne DLleme.! Section, Laboratory of Vectors and Parhot;ell.\', Rocky Moun/ain Labomtories, Ilamilton, Monralla, l-'SA.

(Received 20 luly. 1989; accepted 25 February 1990)

Lyme borreliosis, a spirochetal infection caused by Borrelia burgdorjeri, may t><;,come clinically activ~ after a period of latency in the host. Active cases of Lyme disease may show clinical relapse follo"'ing antibiotic therapy. The latency and relapse phenomena suggest that the Lyme disease spirochete is capable of survival in the host for prolonged periods of time. We studied 63 patients with erythema migran" the pathognomonic cutaneous lesion of Lyme borreliosis. and examined in vitro cultures of biopsie, from the active edge of the erythematous patch. Si~t~en biopsies yielded :;pirocheles after prolonged incubation:; of up to 10.5 months, "Llgg~sting thaI. Borrelia burxdorji>ri may he very slow to divide in eertain situations. Some patients with Lyme borrcliosis may require more than the currently recommended two to three week conr,e of antibio1ic therapy to eradicate strains of the spirochete which grm\' slowly.

Key word:;: Borrelia burgdorjeri; Lyme borrcliosi,; Delayed growth

Introduction

Borrelia burgdorferi, a fastidious spirochetal pathogen, is the etiologic agent of Lyme borreliosis, Since Dr. Willy Burgdorfer's discovery of the spirochete in 1981 (Burgdorfer et aI., 1982), relatively few cases of its isolation from human tissue or body fluid have been reported (Benach et ai., 1983; Chengxu et al., 1988; Hovmark et al., 1986; Huppertz et aI., 1986; Neubert et al.. 1986; pfister et aI., 1984; Preac-Mursic et aI., 1986; Preac-Mursic et aI., 1984: Rawlings et al., 1987; Stanek et al., 1990; Steere et aL 1983; Stiernstedt et aI., 1988). More than 13000 cases of Lymc borreliosis from Lhe United States have been recorded in the Centers for Disease Control Registry since 1982 (Centers for Disease Control, 1989). The infection is encountered by physicians in North America, Europe, the Soviet Union, China, and Australia. In certain humans, B. burgdorferi has caused disease after a period of clinical latency and this has prompted comparisons between this spirochete and Treponema pallidum (Pachner, 1988). A special example of prolonged survival of B. burgdorferi in humans is the condition acrodermatitis chronica atrophicans (ACA), from which skin

Correspondem'f addre,\'s: Dr Alan R MacDonald, Department or Pathology, Southampton Hospital, Southampton, NY 1196S.ITeJephone (516) 2!\3-2600]

OOOl·706X/90/S03.50 (C 1990 Elsevi~r Scienc~ Puhli8h~rs B.V. (Biomedical Divi:;ion)

90

lesions that have been present for 10 years have yielded B. hur!{dorferi in cu lture fro m biopsy materia l (A sbrin k and Hovmark. 19R5). Survival in the host tissue for a decade suggests that B. hurr;dorferi may , in certain circumstances, show ve ry delayed growth in humans. We investigated the possi bili ty that cultiva tion of biopsy ~pecimcn ) rmm erythema migmlls fo r period s of IIp to one year might reveal Slraim of B. burgdcrjeri with 'lery slow growth.

M u eri31s and Methods

From June 1986 to September 19H7, biopsies o f skin from the advancing edge of the centrifugally expa nding erythema mi gra ns lesions were obtained from paTient~ in the Southampton area of Long Island. New York. Biopsies were tran sferred to sterile plastic vial s conta ining 6.0 ml of modified B3rbour-Stoenner-Kelly (RSK) medium. as previously dcscrihed (Ba rbour, 1984; Berger et al., 1985) but lack ing rabbi! serum and gelatin . The biopsies obtained in 1986 were vortexed fo r one minute. len in the culture medium for 24 h, and then removed . 8iopsies in 1987 were o btai ned using identical technique. but in contrast to the 1986 specime ns, they were le ft in the cultu re medium fo r 12 mo nths. Dar kfield exam inatio n o f each cult ure tube bega n 3 weeks after inoc ulation a nd was repeated at mo nthly intervals until spiroc he tes were detected ()T un til 12 months had elap:;ed. All specimens were incubated a t 35 cC fo r the first three weeks and were then maintained at 24 "C for the remainder o f the st udy. Aliquots from positive cultures were examined by sodium dodecyl sulfa te­polyacrylamide gel electrophoresis (SDS·PAGE) (Barbour and Schrumpf, 1986), a nd Western blot Analysis using momx:lona l amibody H5332 which i~ reactive with the outer surface protein A (OspA) unique to B. bllrgdQrj eri (Barbour et aI., 1983).

Results

Spirochetes were recovered after prolonged incubation of skin biopsies from 16 patients with erythema migrans lesionli (Table I). The tim e to detect motile spi rochetes by dark field examination averaged 18 1 days (6 m onths) with a range of 76 319 days (2.5- 10.5 months). However. during 1987 when biopsicd material was left in the culture medium. spirochetes were seen sooner than during 1986 with mean incubation times for detectable growth hei ng 147 days and 208 da ys, respectively.

TABLE 1

Numbt: r of i$OJ:lIion~ of Horre/in imrgliv,/.-,i from hlJm"n 5km biops.ics and illcubauon ~rioo~ for cultures to ~cnmc positive

Incubation (days)

Year Posit ive Contaminated To l ~1 M~an Range --- -'-------

1986 , 2 " -, 2()~ 155- 3\9 1'f10 1 , ., 147 i t. 235 Total " 9 6J 111 1 i6-319

91

Specimens processed in 1986 yielded 9 primary isolates (39%) while in 1987 only 7 isolates were recovered (17%). Spirochetes were not found in any orthe cul1ures that became contaminated. None of the isolates in subculture grew as rapidly as the prototype strain of B. burgdorferi, B31, (American Type Culture Collection F3521O). SDS-PAGE of whole cell Iysates for 8 of the 16 isolates showed protein profiles typical of B. burgdorferi (Fig. 1). Although these eight isolates showed very similar protein profiles, some differences were evident. For example, isolate number 5 had a dominant protein with an apparent molecular mass of 24 kDa which was not as pronounced in any of the other isolates examined. These eight isolates all showed the dominant outer surface protein OspA. which has an apparent molecular mass of approximately 31 k Da. By Western blot analysis, monoclonal antibody H5332 reacted with OspA or the eight isolates (Fig. 2), which along with the SDS-PAGE profiles, identify these isolates as B. burgdorferi. We did not examine the other eight isolates by SDS-PAGE or reactivities with monoclonal antibodies. However, the fact that these other isolates were also cultured in BSK media from active erythema migrans lesions from patients living in an endemic Lyme borrcliosis area, makes it unlikely that these other spirochetes were something other than B. burgdorferi.

s 2 3 4 5 6 7 8 831

92.5 -

66.2 -

45-

31-

14.4 - -';:ii.' Fig. I SDS-PAGE profiles of whole-cell lysate, of Borrelia hurgd"~feri "rt~r prolonged incuhation> from .skin biop,ies taken from erythema migrans lesions or ~ight human patients with Lyme horrelio.sis (lanes I -8) and the prototype strain B31. Arrow indicates position of outer surface protein A. Molecular mas, ,tandards (S) ~rc from Bio-Rad I.aboratories, Richmond, Calif. Gel stained with Coomassie Brilliant Blue.

92

kDa 1 2 3 4 5 6 7 8

31-

hg. 2. The \\'estcrn blot reactivity of isolates 1-8 with monoclonal dnlibody H53J2. ~pe~ili<: ipr <)llter

, urfacc protein A of B. bun.;dork ri_

Discussion

Previous investigators may have underestimated the amount of time that is requIred for certain wild strains of B. burgdorji'ri t.o divide in V1\,0 or in vitro. In the laboratory. strain B31. adapted to in vitro wltiva tioTI, divides rapidly (every 6 to 12 h) (Barbour and H::lYcs. 1986). and has been viewed as a reasonable model for the study of the antimicrobial chemotherapy or Lyme borreliosis and the humoral immune response of various hosts to inl'cction by B. hurgdor[eri. In vitro cultivation of other strains, however, has demonstrated the loss of spirochetal proteins and possible virulence factors, rendering them incapable of producing infection in experimentally inoculated animals (Schwan and Burgdorfer. 1987; Schwan cl aL 19881·

Specics or Borrelia other than B. hurxdor{eri may survivc for remarkably long periods of time in various hosts. Felsenfeld (1971) reviewed reports by several invcstigators showing that Borrelia species may survive in Ornithodoros ticks in the laboratory for 2 to 12 months. Survival of Borrelia species in thc brams of guinea pigs for 1 to :i years without detectable disease in thcse hosts was exploited by investigators earlier in this century (Felsenfc1d, 1971) to preserve strains of Borrcria prior to the discovery by Kelly (1971) of an artifkal medium for in vitro cultivation.

A minimal essential medium for in vitro cultivation of B. hurgdorleri has not been defined. The original formula of Kclly (1971) has been modified several timcs and each modification has been uscd successfully to cultivate B. hurgdorf"cri from human tissue or body fluids. Irrespective of thc BSK modification, however, wild strains of B. hurgdorfi.>ri are difficult to recover and for this reason attempts to culturc thc organism cannot be expected to be a practical diagnostic tool for the practicing physician. BSK medium without rabbit serum has been used (Berger ct aI. , 19H:I: Bergcr et aI. , Ins: Asbrink et aL 1(84) to isolalc B. blirKdorji."i from skin lesions. Johnson et al. (1984) added antibiotics to suppress the growth of nonspirochctal microbes, and Russell C. Johnson (unpublished) has added rcducing substances to enhance the medium. Bovine serum albumin (Fraction V) from commercial sources should be tested each time a new shipment is received because some preparations may be unsuitable for in vitro cultivation or B. hUTxt/offeri and could influence the time for a culturc to show growth. We ehose not to add antibiotics, rabbit serum or a viscosity

9)

enhancer such as gelatin or agarose, none of which is required for the in vitro growth of B. burgdorferi (Barbour and Hayes, 1986). The incubation temperature of 35"'C for the first three weeks was identical to the temperature used by most other workers presented in Table I. However, we realize that the media and cooler temperatures used after 3 weeks in our slUdy may have had some effect on retarding the growth of the frc~hly isolated spirochetes. Yet strain 831 grew well under these conditions. Also, media in which skin biopsies were left produced more rapidly growing spirochetal cultures compared to media in which biopsies remained for only the first 24 h. Therefore, it is unlikely that skin biopsies retarded the growth of B. burgdorferi but may have actuaJly enhanccd the media. Our results also demonstrate that holding cultures for isolation attempts for longer than 3 weeks may be warranted, although we certainly arc not suggesting that such prolonged incubation periods of up to 1 year have any practical role in the diagnosing of active B. hurKdorferi infection.

Aggressive antibiotic therapy of Lyme borreliosis may fail to eradicate the clinical symptoms of the infection (Dattwyler et aI., 1988). Some cases that appear to be treatment failure may actually be due to reinfection. However, at least two well documented cases of Lyme borreliosis have been described which demonstrate that the infection may relapse in spite of parenteral antibiotic therapy administered over a time course which would be expected to kill all spirochetes with a cell division eyele of between 6 to 12 h (Diringcr et aI., 1987; Case report in New Eng. J. Med., 1988). Antibiotics arc only able to kill actively dividing Borrelia. If a borrelial cell does not divide at least once dunng the period of antibiotic therapy, it may persist in the host and produce relapse or recrudescence of disease. Therefore, some cases of Lyme horreliosis may require prolonged periods of antibiotic therapy to influence those cell lines of B. burKdorferi with a very slow cycle of cell division.

Acknowledgements

We thank Merry Schrumpf and Robert Karstens for technical assistance, Gary Hettrick and Robert Evans for photographic work and Betty Kester for preparing the manuscript. Research funds to underwrite this study were generously provided by Southampton Hospital, Southampton, N.Y. and the Neal A. McConnell Foundation.

References

Asbrink, E , H.:derstedt, H., and Hovmark A. (1984) Th~ spiroch~taJ etiology of eryth~ma ehronieum migram Afzclius. Acta [)crm. Venereol. 64, 2YI-295

Ashrink, E. and IIovmark, A. (1985) Successful eultiv3tion of spiro~hete~ from ,\:in lesions of patients with eryth.:ma chrunicum migran8 Afzdius and acrodermatitis ehronica atrophican~. Acta Palho!. Mi<:rob. lmmllnol. Sl:~nd. Se~\. S, 93, J61 163.

Barbour, A.G. (1984) lsoJation and cultivation of Lyme disc<1.se >pirol:hetes. Yale 1. BioI. Med. 57, 521 525.

Barbour, A.G. and Hayes, S.F. (\986) Biology of Borrelia species. Mi<:robio!. Rev. 50, 381-400. Barbour, A.C;. and SchrumpL M.E. (19R6) Polymorphism> of major surface proteins of Borre/ia

hurgJorferi. Zbl. Bakteriol. Hyg. A ~63, R3 91 Barbour, A.G., Te>sier, S.L. and Todd, W.J. (\983) Lyme disease spirochc\~~ and hodid til:k spirol:hetes

share a common surface antigenic determinant defined by a monoclonal anlibody. Infect. Immun. 4\. 795 804.

94

Benaeh, J ,1.. , Bosler. L\{ .. Iidnrclhan, J_P,. C(l l~m :.ln, J. L. , Habicht, G,S" BaSI . T.F .. (;,meron, D,L Ziegler, .1 .1... !larbouf, A.G " Burgdorkr, .... ' .. E<leimall, R. and Kasiow, R.A. (198 3) Spiro~he t e~ isola ted from the blood ,-,1 two p<ltient~ with Lyme di~aM:. -r-;cw Eng. j. M ed. ~08 , 740- 742.

Berger. B,W. , CkmmenM>n. OJ, and Ackermann. A ,B. (1983) Lyme di seast: i> <l spir()c helO~i ~, Am. J_ Dermatopathol. 5. I I [ 124.

Berger. B.W .. Kaplan , MH , Rothcnhcrg, LR. lind Barbour. A,G. /1'11:\51 I>o lalioll ~nd ch~mct.: rilllt i on

o f the Lyme d isease sp irochete from skin o f Illtuen ts w, lh erythema ch rnnicum migrans, 1. A.m . .'\c;,d. Derm . 13. 444-449.

Uurgdo rfcr. W .. Barbo ur. A .G. , J-I aye"l, S I'. " cnach. J .L. . CfUn"~dldl . E. 3m! Da\ t~. J 1'. ( 19K2) Lyme d isease. a IId:.-borno: spi rochclnjlj~ Sciencc 216. U I 7- 1 3 19.

Ca se r.:cord\ of the MIl!loslI chu>dls General Hv:-p' I>l1 (I 1J~~j (Ca se 51-19118) New ~n~ . J . Med . . ~19. 1654 1M2

Centc l"~ fo r Disease Control 11989) Lym~ (li I.C;J$e United States. 1987 ~nd 1988. MMWR .1 ~. M~ 1>72. Chengxu . A. , Yuxin, W. , Yonggou, Z .. Sha,, ~han . W .• Quichcng, Q., Zhi:wt. 5 .. n~you , L . D,)ngquan.

C, Xiaodong. L. ~nd Jienhua. Z. (19!';H) Clinic ~ J manilb lalions and epidcmiolngicaJ char~ c l c ri za t i on or Lyme di l>t'a.~ in La ilin County, Heilongjiang PrtlVill~~, China. Ann, NY Acad, Sci, 539, 302- 313.

D aHwyler, R .J , Halperin . J1. , Volkman. DJ . and LuI'\. lU , (l9!'X) Tn:a tment oflue Lyme bom:l i l)~;' r~mlum j/ed e;lm!'l£l ri~) 11 o f ccftri:txo ne :llId p~njc;lI in . Lanttt I , 119 1 ·1 1')4.

Diringer . M r-.: .. Halperin , 1.1. ~nd D at!\'''yle r, R.J . (1 'i~ 7) Lyme mClli ngocnaplt:, hll~ - I ~po..'fl "I'" se. cn: P'<'n icill in l'C:'I i~tan t C'.J ">C. Art hrit is Rheum 30, 705 71 ll'C .

Fel~nrdd , O . (19 71) Borrelia . Str" illS, \'~Q r:-; . Humnn and Aru mal Bundiusi~. Wll rn:n H. ( ,,,,-'Cn, Inc .. St. Lo)u i~.

Ho.mark. A .•. '\ )bnnk. E. and OIS.-.oIl , I. ( 19116) The spi rochet<ll eliolugy of lymphadcno\1<; lx'.ni,!; II a cUlis ~olit3na . A~ I (' Oo:rm. Venrrcol. fK>, 479-4R4.

Huppo;:rll. . H ,- I. , Slichl-(iroh. '" and Schwan. T. I I ~gti) HMdcrlinc a nt ibody r~>pon~e in Ill!!"'. sl"!lCS <If lymphtX'yt ic mellingili~ due~ nlll mk Olit Ilorrdios is. L~n~cl 2. 146~ 14h'l.

Johnson. S,E .. Klein. Q,C., Sch mid, G.P .. liowcn, 0,5,. Fede)' . J.e and Schlll7C. T (19841 Lyme di~m.c :

a ~le<:ti vc medium for i~olat ion of the > lI ~pe~ t cd etll.lugical agent. a spirochete, J, (' lin . Mi<,·whi"l. 1'1. 81-n

K ~\ly. R. (I \l il), Cu l l i v~ tiun of Borre/l" ke,m,I'i. Science 173. 443- 444. N<:uhcrt . W .. K rum pi tL H.E. <l 1I(t Engl. H (191l6) \tf icrobioloj!:i-="I fi ndi ng$ in <:rrthema (dlfOniclI llll

mi!l'ran~ a rid rdated d i:>ord ~I'S. Z bl. Hak l<:riol. Il yg. A 26 3. 137- 25-:'. . Pac hner. A.R . (1 9881 8"r'1'I;0 blfrgdurier;;n Ihe nervo us s~'St em . the ncw 'grea l imitalor ', Ann. r-.;V A-=ad .

Sci. 5_\9, 56-64 . Pfister, fI .W .. Emha upl, K.M._ P~ac-MUlr.JC, V., Wils~ e. H. and SchicJ7, G . (1984) T he ~"pi rochc\a l

e1 iology "r lymp hocyt ic meningor<l diculi l;~ vf Rann wlI ri h (Bannwanh's syndrume). J :>Icu rol. ~J I. 14 1 144.

Preac-Mll rsh:. V" Wi)~kc, R. a nd Schierz. G, t1 9B6) E\lf<.lpcan Bo"di" i!llrgdor(('ri i:;<" la led from huma ns and tick!; - ~u1tu re conditions and ant ibi o t i~ su>cept;hilit), ZbJ. Bakteriol. Hy!!, !\ 2/'13, ) 12 II~,

PI!:a~ - 'vlur~ic. V .. Wilskc, B .. Schierz. G. PII ~tcr. H.W. and Einhaupl. K (I'lS4) Repealed is<.lJat ion 01' sp; Tllch c tc~ from lh ~ cc rebrospin~l nllid nf a patient with meningor<,d iculiti~ !3.\lll\w,. rt h, Eur. J, elin M iw lr.ini J. 564-565,

Rawlings. J .A .. Fourn i~t, P.\' a nd Tcl to w, G.J. (!987) 1 s< )I~ l io n o f Bur(cfia spin."'h ~I ':'< tr. ,rn pat ients in Tex<l~. J. C hA o M ic r<> . ~}, 1148- 11 50.

Sch"~Jn , T .<;, a nd 8urgdorfer, W . (987) ,\ nligen;( change~ o f Bor refw hurxtliJrjl:fI us a result of ill ";Im cult ure J. Infoct. Dis . l ~, 852- 853.

Schwa n , L G .. BurgdorCer, w . ~nd (laron, c. r. (1988) Changes in infect iyity and phu ll1 id pm fllc of the Lyme d be;ll< >pi rochete, BIJrrelia burgdQTJerl , a$ a rC!lult of in yit ro cultintl iu ll . lufce t. Immun. 56. 1831-1836.

Stand:.. G" Klein. j " Bittner. R. and Cilogar. U, (1 990) I",lal;on or B",r..!w bu'gi/orlff! i'mTJ1 t h~

myocard ium of a patient with longstanding cardlOm~opathy. ~. Eng!. 1, \ted, ,122, 24<j 251 Sltere. A c., (irod7. icki. R .L., Kornblatt , A.N ., Crart. J 1.1 .. Ba rbour. A.0" Rllrgdorfcr. W .. S~hmid. G.P ..

Jnhnson. E. ~nd M~l~wisla. S.E. (1'lH3) The .piro~h~llI l etiology of Lyme dis .. ~ sc , N ...... F.ng. J . \ted, 308. 7.1 3- 740.

Sliernstcdt. G .. Gusmfsson. R .. K <lrl >s<.m. M .. SvcnllngSSOll. B. a nd Sko!denbcrg. B. (19~1\) Clinkal lI1a. n ir~tallons ~,"d d ; agnosi~ o f n~u\"Obom:ho~h . Ann. :-I Y Aca d. Sci. 539. 46 .'i~ .