Journal of Immunological Methods, 149 (1992) 55-62 55 © 1992 Elsevier Science Publishers B.V. All rights reserved 0022-1759/92/$05.00

JIM 06246

Mycoplasma contamination in human leukemia cell lines

II. Elimination with various antibiotics

Cord C. Uphoff, Suzanne M. Gignac and Hans G. Drexler German Collection of Microorganisms and Cell Cultures, Human and Animal Cell Cultures Collection, Braunschweig, Germany

(Received 1 July 1991, accepted 29 November 1991)

19 suspension cell lines were treated with antibiotics for elimination of chronic contamination with mycoplasma. We compared the efficiency, cytotoxicity and cross-resistance of the commercially available antiboitics MRA (Mycoplasma Removal Agent, a quinolone derivative and DNA gyrase inhibitor), Ciprobay (ciprofloxacin, also a quinolone derivative and DNA gyrade inhibitor), and BM-cyclin (a combination of tiamulin, a pleuromutilin derivative, and minocycline, a tetracycline derivative, both inhibitors of protein synthesis on ribosomes). Contaminants were eliminated in all 19 cell lines by BM-Cyclin. Only 74% of the cell lines were cleared of contamination by both MRA and Ciprobay. Successful treatment was monitored by three mycoplasma detection assays. Cross-resistance was noted between MRA and Ciprobay in four of the five cell lines not cleared by either reagent. This resistance could, however, be overcome by consecutive exposure to BM-cyclin. Employed at the recommended concentrations, the antibiotics did not cause marked cytotoxicity, but the growth of the cells was affected to various degrees by some antibiotics. The elimination of mycoplasma from chronically contaminated cell lines is an effective alternative to other treatment protocols, but is cost-intensive and time-consum­ing; lasting damaging effects of the treatments on the eukaryotic cells cannot be excluded. Long-term post-treatment monitoring is mandatory, since contaminants may only be suppressed and then recur.

Key words: Mycoplasma; Cell line; Leukemia; Elimination

Introduction

Mycoplasma infection of continuous cell lines appears to be the major problem of cell culture. Large numbers of cell cultures are infected with various mycoplasma and acholeplasma species and the reported incidence seems to be increas-

Comspondence to: H.G. Drexler, DSM, German Collec­tion of Microorganisms and Cell Cultures, Mascheroder Weg IB, 0-3300 Braunschweig, Germany (Tel.: 49-531-618760; fax 49-531-618718).

ing (Del Giudice et aI., 1984; McGarrity et aI., 1985; Mowles, 1988; Hay et aI., 1989). Mycoplas­mas can alter a great variety of cellular properties and will affect virtually every cell culture parame­ter often leading to experimental artefacts and spurious results.

Because of the high frequency of mycoplasma infections now recognized more easily due to improved detection methods (Uphoff et aI., 1991) and the heightened awareness for the necessity of clean cell lines, attempts are frequently being made to clear infected cells of contamination. Four general procedures have been employed to

56

eliminate mycoplasmas from infected cell cul­tures (Hay et aI., 1989): treatment with animal hyperimmune sera or human serum containing complement (Jeansson et aI., 1985; Nair et aI., 1985; Ziegler-Heitbrock et aI., 1987); passage through nude mice or co-cultivation with mouse macrophages (Van Diggelen et aI., 1977; Schim­melpfeng et aI., 1980); exposure to nucleic acid analogues (Marcus et aI., 1980); and antibiotics (Schmidt et aI., 1984; Borup-Christensen et aI., 1988; Mowles, 1988; Gignac et aI., 1991).

Attempts to eliminate mycoplasma from con­taminated cells should only be considered as a last resort (Hay et aI., 1989). If cells are found to be infected with mycoplasma, the recommended procedure is clearly to discard and to replace them with clean cultures (Del Giudice et aI., 1984). An invaluable or irreplaceable cell line might be worth saving by anti-mycoplasma treat­ment. Irrespective of the choice of method, the treatment protocol plus post-treatment monitor­ing is time-consuming and laborious. An addi­tional possibility is that the cells could undergo a selection process resulting from the potentially toxic conditions, and thus the cured population might differ from the original.

The technically simplest alternative with the most promising results is antibiotic treatment. Tetracycline, kanamycin, lincomycin, ery­thromycin, neomycin, gentamycin, tylosin, minocycline, tiamulin and others have been used alone or in combination with various results (Mc­Garrity et aI., 1985; Hay et aI., 1989), occasionally inducing resistance (Schmidt et aI., 1984).

Recently, new antibiotic reagents, MRA (= mycoplasma removing agent containing a quino­line) and BM-cyclin (a mixture containing a pleu­romutilin derivative and a tetracycline derivative), which are commercially available products specif­ically targeted at mycoplasma infection in cell culture, have met with greater success. A third antibiotic treatment using ciprofloxatin ( '"" Ciprobay) has a great potential for curing my­coplasma-contaminated cell cultures (Mowles, 1988; Schmitt et aI., 1988; Gignac et aI., 1991).

The aim of the present work was to compare these latter three antibiotic treatment protocols with regard to: (j) their effectiveness in eliminat­ing mycoplasma from infected cultures; (ii) any

resistance of the microorganisms and a possible salvage by a second round of treatment with other compounds; (iii) cross-resistance between the reagents and (iv) possible cytotoxic effects towards the cells.

Materials and methods

Cell cultures 19 continuous human leukemia cell lines listed

in Table I were found to be infected with various mycoplasma species. The identification of the various mycoplasma species and the characteris­tics of the cell lines used have been outlined in the accompanying article (Uphoff et aI., 1991). Cells were grown in RPM I 1640 medium (Gibco, Eggenstein, Germany; cat. no. 041-01875) supple­mented with 5-20% (according to cell line) heat­inactivated fetal calf serum (Gibco; cat. no. 011-06290) under standard conditions at 37 0 C in a 5% CO2-humidified atmosphere.

Prior to antibiotic treatment, we attempted to ensure that the cells were in optimal growth conditions. This was often difficult due to massive mycoplasma contamination.

Mycoplasma detection Contamination with mycoplasma was assessed

using three different procedures as described in detail previously (Gignac et aI., 1991; Uphoff et aI., 1991): the classical microbiological cultivation on broth and agar specifically adapted for my­coplasma detection, immunofluorescent staining of mycoplasmal DNA with OAPI, and nucleic acid hybridization of a radioactive probe specific for mycoplasmal ribosomal RNA.

Microbiological broth-agar. Cell culture su­pernatant was incubated with Friis and PH broth for 4-5 days. Broth was then added to Friis- and PH-agar plates, which were then incubated at 37 0 C for up to 4 weeks. Formation of my­coplasma colonies on the solid medium was ex­amined under an inverted microscope (Uphoff et aI., 1991).

DAPI DNA staining. Air-dried cytospin slide preparations were stained with the OAPI (4',6-di­amidino-2-phenylindole-dihydrochloride; Sigma, Oeisenhofen, Germany; cat. no. 0-1388) solution.

Positivity for typical mycoplasma staining around the cell membrane of the eukaryotia cells was evaluated by immunofluorescence microscopy (Uphoff et aI., 1991).

RNA hybridization. Cell culture supernatant was incubated with the radioactive probe of the commercial test kit ('Gen-Probe', San Diego, CA, USA) for 18-20 h. Positivity was examined in a {3 scintillation counter (Uphoff et aI., 1991).

Treatment with antibiotics Mycoplasma-positive cell lines were treated

with the following three antibiotic regimens for 1-3 weeks: mycoplasma removal agent, Ciprobay, and BM-eyelin.

During the treatment, cell lines were kept in 25 cm2 flasks (Nunc, Wiesbaden, Germany; cat. no. 799050) at high density in 10 ml medium. Every second day spent medium was exchanged and antibiotics were added to the fresh medium at the doses indicated. At the end of the treat­ment period cells were centrifuged and resus­pended in fresh medium.

TABLE I

57

After completion of the treatments cells were grown antibiotic-free for at least 2 weeks in order to give resistant mycoplasmas the possibility to mUltiply. All cultures were then tested with the three mycoplasma detection assays described above.

Mycoplasma removal agent. Mycoplasma re­moval agent (MRA; ICN-Flow, Eschwege, Ger­many; cat. no. 30-500-44) was especially devel­oped for use in elimination of mycoplasma in cell cultures. The solution contains a 4-oxo-quinoline-3-carboxyl acid derivative and was added to cell cultures contaminated by mycoplasma at a con­centration of 0.5 J.Lg/ml (= 10 J.LlJmI). Cultures were treated for 8 days (Fig. lA).

Ciprobay Ciprobay (Ciprobay 100; Bayer, Leverkusen,

Germany) is a wide spectrum antibiotic contain­ing ciprofloxacin lactate. Cells were exposed to 10 J.Lg/ml (= 5 J.LlJmO Ciprobay for 14 days (Fig. IBl .

RESULTS OF TREATMENTS WlTIl TIlREE ANTIBIOTIC REGIMENS FOR TIlE EUMINATION OF MYCOPLASMA INFECTION

Cell line c Cell type MRA Ciprobay BM-eyclin M. species C

CrY-I Monocyte R 8 R + M. arginint HDLM-2 Hodgkin cell +b + + M. orale HEL Erythroid cell + + + Unidentified JURKAT Tcell + + + M. orale + M. hyorhinis JVM-3 Bcell + + + A. laidlawii K-562 Erythroid cell + + + M. arginini + M. hyorhinis KARPAS-299 Tcell R + + M. arginini KE-37 Tcell + + + M. arginini + M. fermentans KM-H2 Hogdkin cell + + + M. fermentans L-363 Plasma cell + + + M. orale L-428 Hodgkin cell R R + M. orale

MKB-1 Tcell + R + M. orale ML-2 Monocyte R R + Marginini NAMALWA Bcell + + + A. laidlawii + M. hyorhinis

PI2/ICHIKA WA Tcell + + + M. fermentans

RC-2A Myeloid cell + + + Unidentified

REH Pre B cell + + + M. hominis

SPI-801 Erythroid cell + + + M. orale

SPI-802 Erythroid cell R R + Unidentified

• R: resistance of contaminating mycoplasma(s) to treatment. b +: cell lines cleared of mycoplasma (as assessed by three detection assays). C Details of cell lines and identification of the mycoplasma species have been descnbed previously (Uphoff et al., 1991).

58

BM-eye/in (minoeycline == BM-eyclin 2). Treatment with BM-eyelin inclueed three cyeles. Each cyele grew cells for 3 days in BM-cyelin 1 (at 10 ILg/ml = 4 ILl/mO alternating with 4 days in BM-eyelin 2 (at 5 ILg/ml == 4 ILl/mO-containing medium. On

BM-eyelin (Boehringer Mannheim, Mannheim, Germany; cat. no. 799050) is a combination of two reagents: a pleuromutilin derivative (tiamulin = BM-eyelin 1) and a tetraeycline derivative

A

r Treatment --1- no Antibiotic.

w

o 2 4 6 II 10 12 14 16 III 20 22 24 26 28 30 32 34 36 38 40 42 Day.

t t t t MRA Mycoplalma-Tellt ng -I

B

I- Treatment + no Antibiotics

w

o 2 C 6 • 10 12 Ie 16 II 20 22 2C 26 21 30 32 3C 36 U CO C2 .. C6 41 Day.

"~"~'ll Ciprobay t- Mycoplasma-r-sll ng -I

C

t- Treatment + no Antibiotici

w w w w w w

II II II II II II II II II II II II II II II II II II II II II II II II II II II II II o 2 • • • 10 12 .. It I' 20 22 24 U 21 au U 14 J6 II 40 41 44 .. 41 ID 52 .... D.y.

' ' '''ttl'''' 1 12211221122

BM-Cyelin I-- Mycoplasma-Te,"ng -I Fig. 1. Treatment protocols for MRA (A), Ciprobay (B) or BM-Cyclin (C). MRA was used for 8 days, Ciprobay for 14 days, 8M-Cyelin for 21 days. Antibiotics were given on the days indicated by arrows: MRA at 0.5 ~g/ml, Ciprobay at 10 ~g/ml, 8M-Cyclin 1 at 10 ~lI/ml and 8M-Cyclin 2 at 5 "111m!. Prior to addition of the other antibiotic in the BM-Cyclin protocol, cells were washed (indicated by W) and suspended in fresh medium. At the end of the decontamination period, cells were washed (W) and then grown in antibiotic-free medium. After a minimum of 2 weeks following completion of treatment, 3 mycoplasma detection tests were performed: DAPI DNA staining, RNA hybridization, cultivation in liquid (for 5 days) and solid agar (for 14 days).

Minimum treatment and test period: 41 days (MRA), 48 days (Ciprobay), 55 days (8M-Cyclin).

switching from BM-eyclin 1 to BM-eyclin 2 and vice versa, cells were washed and transferred to completely fresh medium (Fig. lC).

Cytotoxic effects Detrimental effects of the antibiotics on the

growth of cell lines were examined in pilot exper­iments on the mycoplasma-free cell lines REH and PI2/ICHIKA W A. Cell viability was analysed by the trypan blue dye exelusion test. Cell num­bers were determined with a Coulter Counter (Coulter Electronics, Krefeld, Germany).

Results

Influence of antibiotics on cell growth The mycoplasma-free cell lines REH and

PI2/ICHIKAWA were exposed separetely to the four antibiotics contained in MRA, Ciprobay and BM-eyclin in order to test for cytotoxic or cyto­static effects of these reagents on the eukaryotic cells. The viability of the cells remained high (> 90% as assessed by the trypan blue test at 0, 48, 72, 96 and 110 h) in the presence of all four antibiotics excluding any cytotoxicity. With regard to growth-inhibition, MRA had no detrimental effects in this experimental set up; ciprofloxacin and tiamulin (BM-eyelin 1) were e1early inhibitory (34% and 39% grwoth inhibition at day 6, respec­tively), and minocyeline (BM-eyelin 2) was most strongly inhibitive (49% growth inhibition at day 6) (Fig. 2).

However, in cell cultures presumably weak­ened by high levels of mycoplasma infection, ex­posure to the antibiotics led to growth inhibition, deterioration of the cultures and occasionally to the loss of the culture; treatments had to be repeated in the latter cases with new my­coplasma-contaminated aliquots of the same cell lines; MRA (in 1/19 treatments), Ciprobay (1/19) and BM-eyelin (5/19). In fact, it was only at the third attempt that ML-2 could be successfully cleansed and subsequently expanded as a my­coplasma-free culture. In all the above cases, the decontamination and the first week thereafter were the most critical periods (high cell density; frequent medium exchange, high percentage of FCS, etc. were important factors).

59

Cell no. (z 10')

25 untreated

--+-- MRA

20 -+- BM-Cyelln 1

--G- Clprobay

~ BM-eyelln 2

15

10

5

o~~~~--~--~--~--~--~--~ o ~ ~ ~ ~ ~ m ~ ~

lime (hr.)

Fig. 2. Growth-inhibition of mycoplasma-free REH cells ex­posed singly to four different antibiotics at the concentrations

used for mycoplasma eradication.

Treatment with antibiotics The outcomes of the treatments of the 19

mycoplasma-infected cell lines are listed in Table I. Following treatment and an antibiotic-free pe­riod of at least two weeks, any residual my­coplasma contamination in the cell lines was analysed with microbiological broth-agar cultiva­tion, immunofluorescent DAPI DNA staining and the radioactive RNA hybridization assay. While BM-Cyelin e1eansed all cell lines of mycoplasma, 14/19 cell lines (74%) were cured by both MRA and Ciprobay. Four of these cell lines (CTV-l, L-428, ML-2 and SPI-802) were characterized by mycoplasmas resistant to both MRA and Ciprobay. These cell lines were infected with either M arginini or M orale (and one unidenti­fied species). However, other cell lines harboring M. arginini or M orale were cured by MRA and Ciprobay indicating that the resistance of the contaminants might be strain- and not species-re­lated. The species M hyorhinis, M fermentans, M. hominis and A. laidlawii were eliminated by

- all three treatments.

60

Cross-resistance In separate experiments two cell lines (crv-l

and ML-2) were treated sequentially: crv-l was treated first with Ciprobay (resistant), then with MRA (resistant) and finally with BM-eyelin (cured); ML-2 was treated with MRA (resistant), then with Ciprobay (resistant) and thirdly with BM-eyclin (cured). Thus, Ciprobay and MRA could not overcome the resistance to the other of these two reagents while BM-eyelin could be used to rescue MRA- and Ciprobay-resistant cell cul­tures.

Monitoring of antibiotic treatment Progress in the elimination of mycoplasma was

monitored in 2-day intervals by direct DAPI DNA staining in 18/19 Ciprobay-treated cell lines and in 8/19 MRA-exposed cultures. On average, on day 4 or day 6 of Ciprobay treatment (occasion­ally as early as day 2 or as late as day 8) and at day 2 or day 4 of MRA-treatment, the DNA stains were judged to have become clearly nega­tive.

However, in three cases treated with Ciprobay and in one case incubated with MRA, whose DAPI cytospins were thought to be negative dur­ing the treatment, the DAPI test became positive in the post-treatment period; this positivity was confirmed by the other two assays (broth-agar and RNA hybridization). Thus, it appeared that the mycoplasmas had been successfully elimi­nated, but in fact the antibiotics had only sup­pressed the level of infection.

Discussion

The contamination of cell lines by mycoplas­mas remains one of the major problems in cell culture. A simple and efficient method for elear­ing the contaminatants is required, in order to obtain mycoplasma-free cultures. A number of different antibiotics has been used in mycoplas­ma control. The contaminant strains, however, often developed resistance to certain antibiotics (Schmidt et al., 1984). Other antibiotics (for in­stance kanamycin, spectinomycin, lincomycin) were efficient in eliminating mycoplasmas, but

only at concentrations which had detrimental ef­fects on the eukaryotic cells, such as marked eytotoxicity (Schmidt et aI., 1984).

Recently, new antibiotics (distributed as MRA, Ciprobay, and BM-eyelin) have been introduced for the elimination of mycoplasmas from cell cul­tures. We evaluated the efficieney of these reagents in treating a large panel of chronically contaminated cell lines. Experimentally contami­nated cell cultures might not realistically reflect the laboratory situation, since chronic and mas­sive contaminations have a wide range of effects on the cells and certainly result in complex inter­at ions between the mycoplasmas and the cells. We also noted a number of double infections.

Schmidt et a1. (1984) reported that the combi­nation of tiamulin, a pleuromutilin derivative, with minocycline, a tetracyeline derivative, killed the four most common mycoplasma species (M. arginini, M. orale, M. hyorhinis, A. laidlawi;) dur­ing a treatment lasting 6-16 days in 11 cell lines. Neither cell toxicity nor resistance of the my­coplasmas were noted. Both antibiotics inhibit protein synthesis by binding to the 30 Sand 50 S subunits of ribosomes, thereby blocking peptide chain elongation.

Schmitt et a1. (1988) and Mowles (1988) re­ported that ciprofioxacin (= Ciprobay) success­fully eleansed mycoplasma-positive cell cultures in nine (killing seven different species) and 26 attempts, respectively. A recent study docu­mented an efficiency rate of 81 % (17/21) for ciprofioxacin (Gignac et aI., 1991). Ciprofloxacin, a quinolone derivative, inhibits the replication of DNA by interfering with DNA gyrase (topoi­some rase II) (Dabbs et a!., 1987; Drlica et al., 1988).

No treatment results have yet been published on the effectiveness of mycoplasma removal agent (MRA) which is also a quinoione derivative and therefore a gyrase inhibitor.

Here, 19 cell lines, recognized as being my­coplasma-infected by a panel of detection assays (Uphoff et aI., 1991), were treated with MRA, Ciprobay and BM-cyclin. While all 19 cell lines were cured of mycoplasma by BM-eyelin, only 74% of the cell lines were successfully treated by either MRA or Ciprobay. Freedom of infection with mycoplasma in the post-treatment period

was ascertained by the three mycoplasma detec­tion methods found to be most reliable in our hands: DAPI DNA fluorochrome staining, 3H_ labelled DNA-RNA hybridization and microbio­logical broth-agar culturing (Uphoff et aI., 1991).

The mycoplasmas in four cell lines were cross­resistant to both MRA and Ciprobay. This is not surprising when one considers the basic structural similarity of these quinolones. These resistant mycoplasma species were M. arginini and M. orale; but since other cell lines harboring M. arginini and M orale were cured, the resistance appears to be strain- and not species-related. However, subsequent exposure to BM-cyclin fol­lowing unsuccessful treatment with one of the two quinolones could still eliminate the contami­nants.

In pilot experiments on mycoplasma-free cell cultures, treatment with the antibiotics used here had no cytotoxic effect on the eukaryotic cells; a certain degree of growth inhibition was, however, noted. Not unexpectedly, BM-cyclin 2 had the greatest growth-inhibiting effect since tetracycline derivatives inhibit both prokaryotic and eukary­otic ribosomal protein synthesis. The situation in the actual eradication program was somewhat different, since a number of cell lines clearly required more attention than usual in order to re-establish viability and growth, whereas other cell lines died. Cytotoxic and cytostatic effects of the antibiotics might be enhanced by the poor condition of cell cultures commonly found in chronically infected cell lines. This situation is different from that of experimentally contami­nated cell cultures due to the long-term interac­tion of mycoplasmas and cells.

Preliminary tests showed that when MRA and Ciprobay were employed at higher concentra­tions, possibly more effective in cases of resistant contaminants, they had marked cytotoxic effects. Somekh et al. (1989) reported no negative effects of ciprofioxacin on HL-60 leukemia cells at the recommended concentrations, but found growth­inhibition and cytotoxicity at higher values. Schmitt et al. (1988) noted no side effects in cell cultures and proliferation assays using cipro­floxacin in doses up to 2.5 times the usual bac­teriocidal concentration. We did not evaluate whether other cellular parameters were affected

61

by exposure to antibiotics, for instance outgrowth of a selected clone, cytogenetic instability and chromosomal changes, loss of response to growth factors, etc. Obviously any alterations to the cell lines induced by antibiotics are a matter of con­cern and require further detailed studies.

Summary Treatment with antibiotics seems to represent

the most effective and simplest alternative for elimination of mycoplasma from contaminated cell lines. However, the use of uninfected samples of the same cell line and, above all, prevention of contamination should have absolute preference over cleansing of mycoplasma-infected cultures. Since MRA showed the least growth-inhibiting effects combined with the shortest duration of decontamination (although this was at least 6 weeks including the post-treatment and test pe­riod), we prefer to apply MRA as the first-line treatment and then, in the event of resistant contaminants, the more aggressive two-step­three-cycles protocol of BM -eyclin as salvage therapy. Antibiotic eradication of mycoplasmas is costly and time-consuming and cell-damaging side effects cannot be excluded. Long-term monitor­ing to detect residual infection is mandatory. The recommendation that MRA should be used as a permanently preventive measure to avoid my­coplasma contamination must be strongly re­jected. Antibiotics in general and anti-mycop­lasma reagents in particular should not be used as a substitute for good cell culture technique.

References

Borup·Christensen, P., Erb, K. and lensen, I.C. (1988) Curing human hybridomas infected with Mycoplasma hominis. 1. Immunol. Methods 110, 237.

Dabbs, D.I.W., Limb, D.I. and Spencer, R.C. (1987) The new fluoroquinolones: A review. Med. Lab. Sci. 44, 59.

Del Giudice, R.A. and Gardella, R.S. (1984) In: E.M. Levine (Ed.), Uses and Standardization of Vertebrate Cell Cul­tures, Tissue Culture Association, In Vitro, Monograph no. 5 Gaithersburg, p. 104.

Drlica, K. and Franco, R.I. (1988) Inhibitors of DNA topoiso­merases. Biochemistry 27, 2253.

Gignac, S.M., Brauer, S., Hane, B., Quentmeier, H. and Drexler, H.G. (1991) Elimination of mycoplasma from infected leukemia cell lines. Leukemia 5, 162.

62

Hay, R.J., Macy, M.L. and Chen. T.R. (1989) Mycoplasma infection of cultured cells. Nature 339, 487.

Jeansson. S. and Brorson, J.K. (1985) Elimination of my­coplasmas from cell culture using hyperimmune sera. Exp. Cell Res. 161, 181.

Marcus. M., Lavi, U .• Nattenberg, I., Rottem, S. and Markowitz. O. (1980) Selective killing of mycoplasmas from contaminated mammalian cells in cell cultures. Na­ture 285. 659.

McGarrity. G.J .• Sarama, J. and Vanaman, V. (1985) Cell culture techniques. Am. Soc. Microbiol. 51. 170.

Mowles. J.M. (988) The use of ciprofloxacin for the elimina­tion of mycoplasma from naturally infected cell lines. Cytotechnology 1. 355.

Nair, CN. (]985) Elimination of mycoplasma contaminants from cell cultures with animal serum. Proc. Soc. Exp. BioI. Med. 179, 254.

Schimmelpfeng, L.. Langenberg, U. and Peters. J.H. (1980) Macrophages overcome mycoplasma infections of cells in vitro. Nature 285. 661.

Schmidt. J. and Ertle. V. (1984) Elimination of mycoplasma

from cell cultures and establishment of mycoplasma-free cell lines. Exp. Cell. Res. 152. 565.

Schmitt. K.. Diiubener. W., Bitter-Suermann, D. and Haddina. U. (1988) A safe and efficient method for elimination of cell culture mycoplasmas using ciprofloxacin. J. Immunol. Methods 109, 17.

Somekh, E., Dover. D. and Rubinstein, E. (1989) In vitro effect of ciprofloxacin and pefloxacin on growth of normal human hematopoietic progenitor cells and on leukemic cell lines. J. Pharmacol. Exp. Ther. 248, 415.

Uphoff. CC, Gignac, S.M. and Drexler. H.G. (1991) My­coplasma contamination in human leukemia cell lines. I. Comparison of various detection methods. J. Immunol. Methods 149, 43.

Van Diggelen, D., Shin, S. and Phillips, D. (1977) Reduction in cellular tumorigenicity after mycoplasma infection and elimination of mycoplasma from infected cell cultures by passage in nude mice. Cancer Res. 37, 2680.

Ziegler-Heitbrock. H.W.L. and Burger. R. (1987) Rapid re­moval of mycoplasma from cell lines mediated by a direct effect of complement. Exp. Cell. Res. 173, 388.