Immunology Letters 110 (2007) 101–109
Mycoplasma contamination and viral immunomodulatory activity: Dendritic cells open Pandora’s box Marco P. Alves, Carlos P. Carrasco, Carole Balmelli, Nicolas Ruggli, Kenneth C. McCullough, Artur Summerfield ∗ Institute of Virology and Immunoprophylaxis, Sensemattstrasse 293, CH-3147 Mittelh¨ausern, Switzerland Received 21 December 2006; received in revised form 22 March 2007; accepted 11 April 2007 Available online 21 May 2007
Abstract During in vitro investigations on the interaction of classical swine fever virus (CSFV) – an immunosuppressive viral pathogen – with monocytederived dendritic cells (MoDC) a soluble factor with a strong anti-proliferative activity for T lymphocytes was found. This activity, with an inhibitory dilution 50% (ID50 ) of 103 –107 , was induced after virus infection of monocytes differentiating into DC. UV-inactivation of the supernatants and blocking experiments with a monoclonal antibody against the E2 envelope protein of CSFV initially indicated a virus-dependency. However, further investigations including filtration and centrifugation experiments as well as antibiotic treatment demonstrated the involvement of mycoplasma. This was confirmed by a Hoechst 33258 staining, PCR and mycoplasma cultures—Mycoplasma hyorhinis was identified as the contaminant. Elucidation of a mycoplasma presence occurred under conditions in which the original virus stocks prepared in SK6 cells were negative for mycoplasma using the above tests. Moreover, conventional passage of the virus on the SK6 cells used for this purpose did not reveal any mycoplasma. It was the passage of virus in MoDC rather than SK6 cells that was required to expose the contamination. Three passages of the anti-proliferative supernatants on MoDC cultures increased the ID50 103 -fold; only when these MoDC-derived supernatants were employed was the mycoplasma contaminant also detectable on SK6 cells. In conclusion, these data demonstrate that regular testing of cell lines and virus stocks for mycoplasma does not necessarily identify their presence, and that application of passage in MoDC cultures could prove an aid for identifying initially undetectable levels of mycoplasma contamination. © 2007 Elsevier B.V. All rights reserved. Keywords: Dendritic cells; Cell culture quality control; Virus stocks; Mycoplasma contamination
1. Introduction Similar to a number of viruses, classical swine fever virus (CSFV) infects dendritic cells (DC) in vitro and in vivo [1–5]. This creates the environment for potential immunomodulation by the virus. CSFV is a small enveloped RNA virus of the genus Pestivirus in the family Flaviviridae [6], and is responsible for the immunosuppressive disease referred to as classical swine fever (CSF). Typical immunopathological characteristics of CSF are peripheral lymphopenia and thrombocytopenia [2], coagulation disorders and atrophy of the thymus and bone marrow [4,7]. Lymphopenia is caused, at least in part, by apoptosis [3], detectable in uninfected lymphocytes. In addition, viable lym-
∗
Corresponding author. Tel.: +41 31 848 9377; fax: +41 31 848 9222. E-mail address:
[email protected] (A. Summerfield).
0165-2478/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.imlet.2007.04.006
phocytes isolated from CSFV-infected pigs do not respond to mitogen stimulation [3,8–10]. Although the lymphocytes are the major target in terms of immunological dysfuntion, the cells of the myeloid lineage – particularly monocytes, macrophages and DC – are the early targets for virus infection and replication in vivo [11] and in vitro [1–5]. Considering that lymphocytes become dysfunctional without being infected, continuing studies focussed on the capacity of CSFV to modulate DC activity. In the course of these investigations, a potent anti-lymphoproliferative effect was identified in supernatants from virus-infected DC cultures. The characterization of this anti-proliferative effect is the subject matter for the present manuscript. Although this appeared to be a property of certain isolates of CSFV, careful analysis determined that the effect was independent of the virus. The effect was dependent on the presence of mycoplasma contamination. This report demonstrates that despite measures of precaution such as regular
102
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109
screening of cell cultures and virus stocks, mycoplasma contamination may remain low and undetected. When brought into contact with the DC, an augmentation of the influence of the mycoplasma becomes evident, leading to “in vitro artefacts”. Nevertheless, application of DC cultures can have the benefit of exposing such initially undetectable levels of mycoplasma contamination, thus ensuring a more accurate interpretation of results pertaining to viral immunomodulatory effects. 2. Material and methods 2.1. Monoclonal antibodies (mAbs) and flow cytometry mAb against MHC class II (TH16B) was purchased from VMRD (Pullmann, WA, USA) and CD80/86 expression was measured using a human CTLA4-mouse immunoglobulin fusion protein (Alexis, Switzerland). Reactions of the mAbs were revealed using isotype-specific FITC-, RPE- or biotinconjugated anti-mouse immunoglobulin F(ab’)2 fragments (Southern Biotechnology Associates Inc., AL, USA). Biotinylated conjugates were developed with streptavidin-Cy5.5 (Dako, Zug, Switzerland). Acquisition of data was performed with a FACSCalibur flow cytometer and analysis using CellQuest Pro (both Becton Dickinson Bioscience, Basel, Switzerland). 2.2. Cell cultures Monocyte-derived dendritic cells (MoDC) were generated as described previously [27]. Briefly, peripheral blood mononuclear cells (PBMCs) were obtained by density-gradient centrifugation over Ficoll-Paque (1.077 g/l, Amersham Pharmacia Biotech) [28], and CD172a+ (SWC3+ ) monocytes purified using magnetic antibody cell sorting (MACS) (Miltenyi Biotech, Gladbach, Germany) with anti-CD172a monoclonal antibody (clone 74-22-15, ATCC, USA). Monocytes were plated in 6-well plates at a concentration of 106 cells/ml in DMEM (Invitrogen, Basel Switzerland) containing 10% (v/v) porcine serum (Sigma Chemicals, Buchs, Switzerland). Cells were fed at days 0, 2 and 4 with GM-CSF (100 U/ml) and IL-4 (1000 U/ml). After 6 days, adherent and non-adherent cells were harvested. BHK-21 (baby hamster kidney cells), Vero (African green monkey kidney cells) and Mv1Lu (mink lung epithelial cells) cell lines were purchased from ATCC (Manassas, USA) and cultured as recommended. Since receipt from ATCC, all cell lines were routinely screened by PCR and periodically by immunofluorescence tests for freedom from mycoplasma contamination (see below for more details). As an additional precaution, the cells were passaged using trypsin which had been acidified to destroy any mycoplasma present.
2.4. Generation of anti-proliferative supernatants by CSFV infection Freshly isolated CD172a+ monocytes were infected in 6-well plates at a multiplicity of infection (MOI) of 0.1 TCID50 /cell with the CSFV strain Brescia – this virus preparation had been given two passages in the mycoplasma-free SK6 cells in our laboratory. The infected monocytes were then induced to differentiate into DC as described above. DC differentiation was controlled by morphology and phenotype [27]. After 6 days, supernatants were harvested to test for immunomodulatory activity. 2.5. T lymphocyte proliferation assay The mitogen concanavalin A (ConA, 10 g/ml) (Amersham; Piscataway, USA) was used to induce T cell proliferation. To this end, 105 PBMC/well were cultured in RPMI 1640 (Invitrogen) supplemented with 2 mM glutamine (Sigma Chemicals), 0.05 mM 2-mercaptoethanol (Sigma Chemicals) and 10% (v/v) foetal bovine serum (Sigma Chemicals), in a 96-well plate. Dilutions of supernatants from the virus-infected DC were added to certain cultures for analysis of potential immunomodulatory activity. These mitogen-stimulated cells were cultured at 39 ◦ C for 3 days. Proliferation was quantified by addition of 1 Ci [3 H]-thymidine (Moravek Biochemicals Inc., Brea, CA) for the final 18 h. The plates were harvested on to filter mats, and counted in a 1450 MicroBeta TriLux radioactivity counter (Wallac, Turku, Finland). The experiment was performed using triplicate cultures for each supernatant dilution. In order to quantify and compare the activity of different antiproliferative supernatants generated, we defined as the inhibitory dilution 50% (ID50 ) the supernatant dilution which give 50% of inhibition on T lympocytess proliferation compare to medium control. 2.6. UV treatment, filtration and virus blocking For certain experiments, the supernatants from the CSFVinfected DC were UV-inactivated for 15 min at 104 J/m2 with a 6 cm distance to the UV lamp. Other aliquots of the DC-derived supernatants were filtered with VectraSpin micro centrifuge tube filters having a membrane pore size of 200 nm (Whatman, Middlesex, UK). Certain samples were treated to neutralize CSFV infectivity – the mAb HC/TC26 directed against the CSFV E2 structural glycoprotein (kindly provided by Dr. Bommeli, Bommeli Diagnostics AG, Bern, Switzerland) [29] was employed at 10 g/ml for this purpose. 2.7. Mycoplasma-related reagents
2.3. CSFV preparations CSFV strain Brescia (kindly provided by H.J. Thiel, Giessen, Germany) was propagated in the porcine kidney cell line SK6 (kindly provided by M. Pensaert, Gent, Belgium) as previously described [5]. Mock controls were prepared from uninfected cell lysates.
For the generation of mycoplasma-free cell cultures, BMcyclin antibiotics (Roche Diagnostics, Rotkreuz, Switzerland) were added to the cells at final concentrations of 10 g/ml for the pleuromutilin derivative (BM-cyclin 1) and 5 g/ml for the tetracycline derivative (BM-cyclin 2). DNA staining was performed with Hoechst 33258 dye (Invitrogen), to give an esti-
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109
mate mycoplasma DNA presence. More accurate analysis of mycoplasma contamination employed the PCR described elsewhere [30], as well as the VenorGeM mycoplasma detection kit (Minerva Biolabs, Berlin, Germany). Mycoplasma detection by immunofluorescence was performed using Ridascreen kit (R-Biopharm AG, Darmstadt, Germany). Culture of the mycoplasma employed both liquid cultures using mycoplasma broth base (BD Bioscience, Basel, Switzerland) and solid agar cultures. Identification of the mycoplasma species present was performed by external experts (Mycoplasma Experience, Surrey, UK). 3. Results 3.1. Generation of supernatants from CSFV-infected MoDC cultures for anti-proliferation studies Initially, the aim of the present work had been to characterize CSFV interaction with DC, towards understanding how the virus influences DC differentiation. Accordingly, the influence of CSFV infection on DC maturation was analysed in terms of MHC class II and CD80/86 expression by flow cytometry. No significant changes in surface markers expression were observable when CSFV-infected cultures were
103
compared with mock-treated cultures. This was the case with both MoDC infected as either differentiated (Fig. 1A and C) or at the monocytic precursor state (Fig. 1B and D). Interestingly, the supernatants from the infected DC cultures at their precursor state possessed a strong anti-proliferative activity for concanavalin A (ConA)-activated T lymphocytes. Dilutions of the supernatants up to 10−3 almost abrogated the lymphoproliferation; only at a 10−5 dilution were the levels similar to those of mock controls (Fig. 2A). Induction of this antiproliferative activity was observed after CSFV infection of monocyte-derived DC (MoDC) precursors, but not by infecting SK6 cells with the same CSFV preparation. Apparently one passage on DC is required to generate the anti-proliferative supernatants (Fig. 2B). The mock (uninfected) controls did not provide any anti-proliferative activity, indicating a CSFVdependent effect. In order to confirm the importance of the virus for this anti-proliferative activity, virus neutralization experiments we performed using the mAb HC26 against the viral structural protein E2. Addition of the mAb at 10 g/ml restored the ConA-driven T cell proliferation (Fig. 2C). UVlight treatment of the supernatants – under conditions known to neutralize virus infectivity – also reversed the anti-proliferative activity of supernatants from CSFV-infected DC cultures (Fig. 2D).
Fig. 1. Influence of CSFV infection on MHC class II and CD80/86 expression on differentiated MoDC and DC precursors (monocytes). The cells were treated in a similar manner with mock (dotted line histogram) or with CSFV (bold line histogram) at an MOI 1 TCID50 /cell. In each case, conjugate controls were used (grey-filled histogram). DC were collected, and the expression of cell surface MHC class II and CD80/86 measured by flow cytometry. Data shown are representative of four independent experiments.
104
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109
Fig. 2. Characterization of CSFV-infected DC-derived supernatants inhibiting lymphocyte proliferation. Monocytes were either mock-treated or CSFV-infected and induced to differentiate into DC in a 6-day culture with GM-CSF and IL-4. The culture supernatants were harvested and titrated on PBMC in a ConA-induced proliferation assay (A). In (B), SK6 cells were either mock-treated or CSFV-infected and cultured for 4 days before harvesting of the supernatants, which were then titrated on PBMC in a ConA-induced proliferation assay. In (C), MoDC-derived supernatants generated as for (A) were diluted 103 times and neutralized with anti-E2 (HC26) (1 h, 37 ◦ C) before testing in the proliferation assay. In (D) and (E), CSFV or mock-infected MoDC-derived supernatants were either UV-inactivated (D; “CSFV UV” or “mock UV”) or filtrated on a 200 nm cut-off filter (E; “CSFV F” or “mock F”) and titrated in the proliferation assay. In (F), the MoDC-derived supernatants were centrifuged for 20 min at 15,000 × g, and the pellet (“P”) and supernatant (“S”) were separated and titrated in the proliferation assay. Before titration, pellet was resuspended in medium on same volume as supernatant. The mean and S.D. of triplicates are displayed. Data are representative of six independent experiments.
3.2. Particulate nature of the anti-proliferative agent The observation that the anti-proliferative activity was obtained only with CSFV-infected DC cultures, but not with infected SK6 cells, pointed to the involvement of an element produced by the DC cultures. One possibility is the production of a cytokine, while a second would be the presence of a particulate agent. Further insight into the nature of the anti-proliferative effect was obtained by passing supernatants from infected DC
cultures through a 200 nm cut-off filter. As expected, the CSFV titre was not influenced by this filtration (CSFV size ∼60 nm). In contrast, the anti-proliferative effect for T cell lymphoproliferation was totally lost (Fig. 2E). These results pointed on the presence of a relatively large agent – certainly larger that a cytokine or the CSFV itself. This was confirmed by centrifugation of the supernatants at 15,000 × g for 20 min, which reduced the 50% inhibitory dose (ID50 ) in the supernatants by over 100-fold – the bulk of the
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109
105
anti-proliferative effect was found to be in the pellet (Fig. 2F). Indeed, the effect was enhanced in the pellet. When a 1:100 dilution of the pellet was mixed with mock supernatants, the anti-proliferative activity was similar to that of the original supernatants from the CSFV-infected DC cultures (data not shown). 3.3. Monocytic cell cultures in contrast to SK6 augment the concentration of the anti-proliferative factor The above results demonstrated that a particulate agent >200 nm in size was involved in the anti-proliferative effect. In order to determine if this agent was a living entity, attempts were made to amplify its presence. Accordingly, the anti-proliferative supernatants were “passaged” three times in monocytes, which were then induced to differentiate into DC. With each passage, a 10–100-fold increase of the ID50 was noted, attaining between 107 and 108 ID50 after the third passage. In all cases, the anti-proliferative activity remained associated with the more particulate fraction retained by a 200 nm filter (as exemplified in Fig. 2E). Considering that the titres of CSFV remained unchanged (∼105 tissue culture infectious dose/ml (TCID50 /ml)), it appeared that a second replicating agent was present and involved in elaborating the anti-proliferative activity. This second replicating agent is likely to be the major perpetrator of the anti-proliferative activity – the activity was present at the dilutions >10−7 in which no infectious CSFV was present (Fig. 3A). Interestingly this increase of activity after DC passage of anti-proliferative supernatants was not observed by passaging it on SK6 cells. Although an activity could be detected in the SK6 cell supernatants shown in Fig. 3B, the level of this activity remained constant with passage. This demonstrates that the MoDC cultures were selectively enhancing the replication of the anti-proliferative agent. 3.4. Identification of the causative agent for the anti-proliferative effects Considering the above results and the apparent size and replicative ability of the anti-proliferative agent, it was possible that mycoplasma were involved. Initially, the CSFV stock used for the infections had been tested for mycoplasma contamination by PCR, immunoflorescence, DNA staining (using the Hoescht 33258 stain) and direct mycoplasma culture isolation. None of these tests were able to show the presence of any mycoplasma in the virus stock. Moreover, the SK6 cell cultures were routinely screened that they were remaining free of mycoplasma contamination; the MoDC were also screened and found to be free. Despite the apparent absence of mycoplasma from the virus stocks, it was considered prudent to control the supernatants from CSFV-infected MoDC which were presenting the antiproliferative effect. Consequently, the active supernatants as well as the CSFV-infected MoDC cultures were tested by the PCR and immunofluorescence tests described in Section 2 for the presence of mycoplasma. Table 1 shows that depending on the
Fig. 3. Monocytic cell cultures in contrast to SK6 augment the concentration of the anti-proliferative factor. In (A), CSFV or mock MoDC-generated supernatants were passaged three times on freshly prepared MoDC cultures infected at their precursor state. After each passage, the supernatants were titrated on PBMC in a ConA-induced proliferation assay. In (B), MoDC-derived anti-proliferative supernatants obtained after the first DC passage were further passaged three times on SK6 cells and tested in the proliferation assay.
method, mycoplasma presence was detected, but there was not an absolute link between this and an anti-proliferative character. When the DNA present in the cultures was stained with Hoechst 33258, a fluorescence image typical for the presence of mycoplasma was found in the CSFV-infected MoDC culTable 1 Comparison of mycoplasma detection methods Supernatant sample
Activity
PCR
IF
Hoechst
Positive control CSFV stock Mock stock CSFV stock 1 × SK6 CSFV stock 1 × DC CSFV stock 3 × DC
− − − + +
+ − − − ± +
+ − − − − ND
+ − − − + +
PCR, immunofluorescence (“IF”) and Hoechst DNA staining (“Hoechst”) methods were used for mycoplasma detection and correlated with anti-proliferative activity (“Activity”) on several samples (“Supernatant sample”). The positive control is a BHK-21 cell line mycoplasma positive used as a reference during regular testing; CSFV and mock stocks are the original SK6 preparations used during the present study. The CSFV stock was “passaged” one time on SK6 (“CSFV stock 1 × SK6”) or one to three times on DC (“CSFV stock 1 × DC”; “CSFV stock 3 × DC”). Data are representative of six independent supernatants for each sample showed; ±: two out of six supernatants were positive; ND: not done.
106
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109
tures from which anti-proliferative supernatants were obtained (data not shown). The final confirmation of the presence of mycoplasma employed liquid and solid mycoplasma culture systems, inoculated with the anti-proliferative supernatants. Very dark colonies typical of Mycoplasma hyorhinis were obtained (Fig. 4A). The identity of this isolate was confirmed by disc growth inhibition using hyperimmune rabbit antiserum (experiments performed by Mycoplasma Experience Ltd.; data not shown). In order to demonstrate that this mycoplasma presence was related to the anti-proliferative activity, we passaged MoDCderived supernatants on MoDC, SK6 and PK15 cells in the presence of specific antibiotics against mycoplasma (BMcyclins). The anti-proliferative effect was indeed lost when the passage was performed in the presence of the BM-cyclins, regardless of the cell type used (Fig. 4B).
3.5. Anti-proliferative activity of CSFV-infected MoDC supernatants on non-porcine cells The results to date were demonstrating an anti-proliferative activity for porcine lymphocytes. As an extension to this characteristic, the host range of the anti-proliferative agent activity was analysed. For this, supernatants from CSFV-infected MoDC cultures were analysed on the hamster kidney cell line BHK-21, the monkey kidney cell line Vero and the mink epithelial cell line Mv1Lu. In comparison with the inhibitory effect for porcine lymphocyte proliferation, it was seen that the supernatants were also effective on cells from these other species (Fig. 5). These results demonstrate that the anti-proliferative effect was active on a variety of cell types – not just lymphocytes – and with a variety of species. 4. Discussion
Fig. 4. A mycoplasma contamination is responsible for the anti-proliferative activity. In (A), a microphograph of an agar plate with mycoplasma colonies is shown. This was obtained by inoculation of 2 ml mycoplasma broth medium with 100 l of MoDC-generated anti-proliferative supernatant. After 2 weeks of incubation at 37 ◦ C, a mycoplasma agar medium plate was inoculated with liquid broth medium culture diluted 104 times. Agar plate was microscopically analysed after 3 weeks of culture at 37 ◦ C under anaerobic conditions. The bar represents 0.1 mm. In (B), the anti-proliferative supernatants were passaged once in the presence of BM-cyclins antibiotics on MoDC, SK6 or PK15 cells. After 4 days culture, obtained supernatants were assayed in a ConA-induced PBMC proliferation assay (C). The mean and S.D. of triplicates are displayed. Data are representative of three independent experiments.
The present work was initiated due to the observation that CSFV-infected monocyte cultures induced to differentiate into DC yielded a factor with potent anti-proliferative activity on T lymphocytes. Assays using mitogen-stimulated T cell proliferation showed that these supernatants had an ID50 between 104 and 107 . Interestingly, no sign of cytotoxicity was observed in the inhibited cultures, but cell counts were reduced (data not shown) indicating a reduction in cellular proliferation. Of major importance was the observation that the same virus stock passaged on SK6 cells did not induce this activity, indicating an essential role for monocytic cells in the generation of the highly active anti-proliferative supernatants. At first, these results appeared to explain the anergy found with T cells isolated from CSFV-infected animals. Moreover, potential mycoplasma contamination was not initially of concern considering that our SK6 cells and the SK6 cell-passaged virus stocks were apparently free of mycoplasma – based on regular screening by PCR and immunofluorescence. In addition, virus neutralization using a mAb against the viral E2 glycoprotein blunted the activity of anti-proliferative supernatants. It was only after the filtration and centrifugation experiments that the anti-proliferative agent was seen to be particulate and too large for CSFV. How the anti-E2 mAb was able to impede this agent, which we now know to be mycoplasma, is unclear. One possibility is that the agent was closely associated with the CSFV, and its interaction with the monocytic cells required “help” from the virus. If this were the case, then once the agent had established itself in the DC cultures, further propagation and maintenance of its presence was independent of the CSFV. Indeed, no mycoplasma were detected in the original SK6 cell-cultured virus stocks, and no sign of mycoplasma was found upon passage of the virus in the SK6 cells. Only when the virus had been passaged in the MoDC was it possible to detect the mycoplasma, and this now maintained its presence upon further passage in the SK6 cells. Considering the filtration and centrifugation analyses, the suspicion turned back to a mycoplasma contamination, even though the original tests on the virus stocks had proven negative by PCR and immunofluorescence tests. These tests did show the presence of mycoplasma in the anti-proliferative super-
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109
107
Fig. 5. Activity of DC-derived supernatants for cells of several species. CSFV-infected or mock-treated supernatants from MoDC cultures were tested for antiproliferative activity on ConA-acitivated human PBMC, BHK-21 cells, Vero cells, or Mv1Lu cells. The mean and S.D. of triplicates are displayed.
natants from infected DC cutlures, but the mycoplasma presence was inconsistently linked to the anti-proliferative activity. Consequently a third test was employed – staining of the cultures with Hoechst 33258 DNA stain. This identified what appeared to be mycoplasma in all anti-proliferative supernatants. Presence of the mycoplasma was confirmed by isolation in specific mycoplasma media. The ability of the mycoplasma to self-propagate in the DC cultures was obtained through dilution of the anti-proliferative supernatants to the point where CSFV was no longer detectable, but the mycoplasma were still present. This procedure generated CSFV-free supernatants retaining anti-proliferative activity, demonstrating that the anti-proliferative characteristics were indeed due to the mycoplasma. The final proof of the relationship between the mycoplasma contamination and the anti-proliferative activity was obtained with the BM-cyclin treatment of the DC cultures before and during infection. This process abrogated the anti-proliferative activity, but did not influence the capacity of CSFV to replicate in the DC. The present studies highlight once again the major problem posed by mycoplasma contamination for laboratories using cell culture [12–15]. Their small size – 0.2 and 2 m – excludes their removal from virus stocks by filtration, and their selfreplicating capacity allows them to exist independently once established in a culture. They have relatively slow growth profiles, which can result in them remaining initially undetected, as indeed occurred in the present work with the original CSFV stocks. Moreover, antibiotics such as penicillin are ineffective at the standard cell culture concentrations used, and as extracellular
parasites mycoplasma often attach to the cell plasma membrane, rendering their removal difficult by physical means. The most common contaminating cell cultures are Acholeplasma laidlawii, Mycoplasma arginini, M. fermentans, M. hyorhinis and Mycoplasma orale [16–18]. For the detection of mycoplasmas, it is recommended to use at least two different techniques such as mycoplasma DNA staining on the surface of infected cells by Hoechst 33258, PCR, immunofluorescence or mycoplasma cultures [19–21]. Mycoplasma contamination has been reported to interfere with T cell proliferation [22–25]. Related to this, the immunosuppressive activity of human cytomegalovirus infected monocytic cells was due to a M. hyorhinis contamination originating from the virus stock [26]. The present work demonstrates the danger of interpreting virus-induced immunomodulatory activity. Our CSFV stocks were considered to be free of mycoplasma – based on the use of PCR and immunofluorescence tests. Moreover, serial passage of the virus in SK6 cells did not reveal any mycoplasma. It was only when the virus was propagated in the DC cultures that the mycoplasma became apparent. Interestingly, once perceptible in the DC cultures, the mycoplasma could now be propagated in the SK6 cell cultures. This would imply that the MoDC are a particularly sensitive cell system for promoting the expansion of mycoplasma contamination, especially from virus stocks in which there is no evidence of the mycoplasma presence. The scheme in Fig. 6 is based on retrospective analyses of our virus stocks and cell culture supernatants, as presented in the present paper. Not only did passaging of the virus on MoDC permit the mycoplasma
108
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109
Fig. 6. Flow chart of contamination history scheme showing the summary of history contamination. The activity of supernatants (“activity”) generated after one to three passages on MoDC and/or SK6 cell cultures is indicated together with mycoplasma contamination (“myco”). The latter was based on analysed by Hoechst 33258 staining, PCR and mycoplasma cultures.
to become detectable, the mycoplasma was now displaying its anti-proliferative capacity for T lymphocytes, an activity which was further amplified by serial passage in the in MoDC cultures. These characteristics were not revealed by passage in the SK6 cells conventionally employed for growing CSFV stocks. As far as we are aware, this is the first report showing an active role of monocytic cells in the expansion of a mycoplasma contamination originating from an undetected contamination in a virus stock. We, therefore, conclude that monocytic cell cultures can preferentially support mycoplasma replication despite the potential antimicrobial activities of these cells. Taken together, the data demonstrate that regular testing of cell lines and virus stocks for mycoplasma does not guarantee elimination of the problem, but application of monocytic cells can prove an aid to detection of the contaminant. Moreover, one has to pay particular attention to interpreting results on virus interactions with monocytic cell cultures – including dendritic cells and macrophages – due to the potential influence of initially undetectable mycoplasma contaminations. Acknowledgements This work was supported by the Swiss Federal Office for Education and Science (project 3100-068237/1). We are very grateful for the help provided by David and Helena Windsor from Mycoplasma Experience Ltd. with the mycoplasma cultures and the identification of the mycoplasma species, to Daniel Brechb¨uhl for care of the blood donor animals and blood preparation, to Heidi Gerber for preparation of the mAb and SK6 cells and to Luzia Lui for regular controlling of the cells lines for mycoplasma contamination. References [1] Carrasco CP, Rigden RC, Vincent IE, Balmelli C, Ceppi M, Bauhofer O, et al. Interaction of classical swine fever virus with dendritic cells. J Gen Virol 2004;85:1633–41. [2] Summerfield A, Hofmann MA, McCullough KC. Low density blood granulocytic cells induced during classical swine fever are targets for virus infection. Vet Immunol Immunopathol 1998;63:289–301.
[3] Summerfield A, Knotig SM, McCullough KC. Lymphocyte apoptosis during classical swine fever: implication of activation-induced cell death. J Virol 1998;72:1853–61. [4] Summerfield A, Knoetig SM, Tschudin R, McCullough KC. Pathogenesis of granulocytopenia and bone marrow atrophy during classical swine fever involves apoptosis and necrosis of uninfected cells. Virology 2000;272:50–60. [5] Knoetig SM, Summerfield A, Spagnuolo-Weaver M, McCullough KC. Immunopathogenesis of classical swine fever: role of monocytic cells. Immunology 1999;97:359–66. [6] Thiel HJ, Plagemann PGW, Moennig V, Fields BN, Knipe DM, Howley PM. Pestiviruses. In: Fields virology. Philadelphia: Lippincott-Raven Publishers; 1996. pp. 1059–1073. [7] Summerfield A, Zingle K, Inumaru S, McCullough KC. Induction of apoptosis in bone marrow neutrophil-lineage cells by classical swine fever virus. J Gen Virol 2001;82:1309–18. [8] Charley B, Corthier G, Houdayer M, Rouze: P. Immunologic abnormalities in pigs with hog cholera virus infection (author’s transl). Ann Rech Vet 1980;11:27–33. [9] Van Oirschot JT, De Jong D, Huffels ND. Effect of infections with swine fever virus on immune functions. II. Lymphocyte response to mitogens and enumeration of lymphocyte subpopulations. Vet Microbiol 1983;8: 81–95. [10] Pauly T, Konig M, Thiel HJ, Saalmuller A. Infection with classical swine fever virus: effects on phenotype and immune responsiveness of porcine T lymphocytes. J Gen Virol 1998;79(Pt 1):31–40. [11] Ressang AA. Studies on the pathogenesis of hog cholera. II. Virus distribution in tissue and the morphology of the immune response. Zentralbl Veterinarmed [B] 1973;20:272–88. [12] Hay RJ, Macy ML, Chen TR. Mycoplasma infection of cultured cells. Nature 1989;339:487–8. [13] Rottem S, Barile MF. Beware of mycoplasmas. Trends Biotechnol 1993;11:143–51. [14] Lundin DJ. Mycoplasmal contamination of cell cultures. Am Biotechnol Lab 1994;12:42. [15] Langdon SP. Cell culture contamination: an overview. Methods Mol Med 2004;88:309–17. [16] Mirjalili A, Parmoor E, Moradi Bidhendi S, Sarkari B. Microbial contamination of cell cultures: a 2 years study. Biologicals 2005;33:81–5. [17] Jung H, Wang SY, Yang IW, Hsueh DW, Yang WJ, Wang TH, et al. Detection and treatment of mycoplasma contamination in cultured cells. Chang Gung Med J 2003;26:250–8. [18] Uphoff CC, Drexler HG. Comparative PCR analysis for detection of mycoplasma infections in continuous cell lines. In Vitro Cell Dev Biol Anim 2002;38:79–85. [19] Sung H, Kang SH, Bae YJ, Hong JT, Chung YB, Lee CK, et al. PCR-based detection of Mycoplasma species. J Microbiol 2006;44:42–9.
M.P. Alves et al. / Immunology Letters 110 (2007) 101–109 [20] Uphoff CC, Drexler HG. Detection of mycoplasma contaminations. Methods Mol Biol 2005;290:13–23. [21] Benisheva T, Sovova V, Ivanov I, Opalchenova G. Comparison of methods used for detection of mycoplasma contamination in cell cultures, sera and live-virus vaccines. Folia Biol (Praha) 1993;39:270–6. [22] Callewaert DM, Kaplan J, Peterson Jr WD, Lightbody JJ. Suppression of lymphocyte activation by a factor produced by Mycoplasma arginini. J Immunol 1975;115:1662–4. [23] Proust JJ, Buchholz MA, Nordin AA. A “lymphokine-like” soluble product that induces proliferation and maturation of B cells appears in the serumfree supernatant of a T cell hybridoma as a consequence of mycoplasmal contamination. J Immunol 1985;134:390–6. [24] Ruuth E, Ranby M, Friedrich B, Persson H, Goustin A, Leanderson T, et al. Mycoplasma mimicry of lymphokine activity in T-cell lines. Scand J Immunol 1985;21:593–600. [25] Mizushima Y, Quintans J, Cohen EP. Stimulation of lymphoid cell proliferation by Mycoplasma orale, a common cell culture contaminant. Infect Immun 1985;50:636–40.
109
[26] Scott DM, Rodgers BC, Freeke C, Buiter J, Sissons JG. Human cytomegalovirus and monocytes: limited infection and negligible immunosuppression in normal mononuclear cells infected in vitro with mycoplasma-free virus strains. J Gen Virol 1989;70(Pt 3): 685–94. [27] Carrasco CP, Rigden RC, Schaffner R, Gerber H, Neuhaus V, Inumaru S, et al. Porcine dendritic cells generated in vitro: morphological, phenotypic and functional properties. Immunology 2001;104:175–84. [28] McCullough KC, Schaffner R, Fraefel W, Kihm U. The relative density of CD44-positive porcine monocytic cell populations varies between isolations and upon culture and influences susceptibility to infection by African swine fever virus. Immunol Lett 1993;37:83–90. [29] Greiser-Wilke I, Moennig V, Coulibaly CO, Dahle J, Leder L, Liess B. Identification of conserved epitopes on a hog cholera virus protein. Arch Virol 1990;111:213–25. [30] Teyssou R, Poutiers F, Saillard C, Grau O, Laigret F, Bove JM, et al. Detection of mollicute contamination in cell cultures by 16S rDNA amplification. Mol Cell Probes 1993;7:209–16.
