Spread and control of mycoplasmal infection of cell cultures

Summary Environmental sampling was performed during trypsinization and passage of 3T-6 cell cultures that contained a mean of 4.3×10⁷ colony forming units (CFU) per ml supernatant ofA. laidlawii. The lip of the culture flask and the outside of the used pipet were always heavily contaminated. The outside of the culture flask (3/7), the work surface (8/12) and the outside of a pan of disinfectant (4/5) were regularly contaminated with mycoplasmas. Airborne mycoplasmas were detected eight of 32 times (25%) by settling plates; simultaneous forced-air samplers by two different methods were always negative. The technician’s hands were contaminated two of 15 samples. When hands were contaminated, more contamination was detected in the environment. Droplets ofA. laidlawii andM. orale inoculated onto work surfaces survived drying for a minimum of 3 days, even in laminar airflow cabinets. Twenty-five of 31 (80.6%) cell culture technicians carriedM. salivarium in their throats; only two carriedM. orale. It is concluded that mycoplasma-infected cultures are the most common source of further infection. Recommendations for prevention and control of mycoplasmal infection are listed. These studies were supported in part by Contract No. 1-GM-2112 from the National Institute of General Medical Sciences, Contract No. 1-CB-23868 from the National Cancer Institute, General Research Support Grant 5-S01-RRO5582 from Research Resources, National Institutes of Health, and by a Grant-in-Aid from the State of New Jersey.     

Microbiological cultivation ofMycoplasma hyorhinis from cell cultures

The failure of many cell culture isolates ofMycoplasma hyorhinis to grow on microbiological media has stressed the need for alternate assays to detect these organisms. The use of freshly prepared yeast extract in mycoplasmal media together with incubation in 5% CO₂/air successfully detectedM. hyorhinis in 12 of 12 infected cultures. These were not detected by the use of conventional mycoplasmal media using aerobic or anaerobic incubation. This assay may also be helpful in detection of other mycoplasmal species commonly isolated from cell cultures. This study was supported by grant AI 15748 from NIH. We thank Ronnie Vanaman and Judi Sarama for their technical assistance.     

Comparative studies to determine the efficiency of 6 methylpurine deoxyriboside to detect cell culture mycoplasmas

Summary Studies were performed to compare three methods to detect mycoplasmal infection of cell cultures. The methods included microbiological assay by inoculation into broth and onto agar with anaerobic incubation, fluorescent DNA staining by Hoechst 33258, and mycoplasmal mediated cytotoxicity by 6 methylpurine deoxyriboside (6MPDR). Fluorescent DNA staining and 6MPDR assays were performed in an indicator cell culture system. A total of 2589 cell cultures were assayed. Mycoplasmas were detected in 174, an incidence of 6.7%. Species isolated were:Acholeplasma laidlawii, Mycoplasma orale, M. arginini, M. hyorhinis, M. fermentans, M. pirum, and M. pneumoniae. In separate studies, 6MPDR also detected infection withSpiroplasma mirum when this organisms was deliberately inoculated into cell cultures. The efficiencies of microbiological testing, fluorescent DNA assays, and 6MPDR were 43.1, 98. 8, and 97.1%, respectively. The work was supported by grant AI-15748 from the National Institutes of Health, Bethesda, MD     

Effect of Mycoplasmas on Apoptosis of 32D Cells Is Species-Dependent

We previously showed that mycoplasmal infection effectively prevented apoptosis of infected cells, whereas other researchers have indicated that mycoplasmal infection promoted apoptosis. To understand the mechanism underlying this discrepancy, five different species of mycoplasmas were investigated for their effects on apoptosis of interleukin (IL)-3–dependent 32D cells. Results revealed that Mycoplasma fermentans and M. penetrans effectively supported continuous growth of 32D cells after IL-3 withdrawal. M. fermentans was more potent than M. penetrans. This effect was achieved by way of preventing apoptosis and stimulating cell proliferation. On the contrary, M. hominis and M. salivarium accelerated apoptosis of 32D cells. M. genitalium had no significant effect on apoptosis. The RNase protection assay indicated that the proapoptotic and antiapoptotic mycoplasmas altered the expression of major apoptosis regulatory genes differently. The difference in apoptosis regulatory gene expression induced by different species of mycoplasmas might be accountable for their effects on host cell apoptosis.     

Pleuropneumonialike organisms in tissue cultures

Of 55 continuous cell lines 32 gave growth of P.P.L.O. whereas 26 primary cell cultures were free from this contamination.Biochemical and serological typing proved that 31 of these 32 P.P.L.O. wereMycoplasma hominis I. One strain was identical with a recently described oralMycoplasma.It was demonstrated that insufficiently rigorous techniques tend to cause spreading of P.P.L.O. in tissue culture laboratories.There was no indication that either the sera or other ingredients of the media used might have introduced this contamination. HeLa cells, however, probably are the source.The minor differences between genital strains propagated in the laboratory, and tissue culture strains, are probably due to differences between the two media.By treatment of a contaminated cell line with serum againstM. hominis I a double infection with P.P.L.O. could be demonstrated. The cells were freed from the remaining P.P.L.O. by treatment with the serum against this strain.Contamination of cell lines withM. hominis I did not affect the growth rate of the wild poliovirus I strains tested, nor that of a Sabin type I strain. M. fermentans grows well in tissue cultures but has no cytopathic effect.M. salivarium cannot be propagated in ordinary tissue cultures unless Fildes extract, which contains catalase, is added. In cultures with this extractM. salivarium has a cytopathic effect.     

Establishment and characterization of immortalized cell lines from rat parotid glands

Summary This study reports for the first time the establishment of immortalized cell lines from normal adult rat parotid glands. The freshly prepared cellular clumps obtained from parotid glands of isoproterenol-treated rats were incubated in 0.2% trypsin solution without EDTA. These clumps were transfected with plasmid vectors pSV ₃ ^neo and pSV ₅ ^neo by electroporation and calcium phosphate-Co-DNA-precipitation techniques. The untransfected and transfected cellular clumps were plated in precoated dishes containing modified MCDB-153 medium. Epithelial cells grew from the clumps that were attached. All epithelial cells from untransfected culture died within 6 to 8 wk. Two cell lines which were isolated from transfected cultures subsequently grew on regular tissue culture dishes. One of them, which was isolated from pSV ₅ ^neo transfected cultures, exhibited non-epithelial cell morphology, but at confluency, many cells mature to acinar-like cells containing numerous granules. The other cell line (2RS), which was isolated from pSV ₃ ^neo transfected culture, contained cells of non-epithelial and epithelial morphology. During the initial phase of the growth, MCDB-153 medium was essential; however, at a later time, RPMI medium was better than MCDB-153 or F12 medium for maintaining morphology and growth of these cells. The immortalized cells grew in RPMI with a doubling time of about 25 h, synthesize T-antigen,α-amylase mRNAs of 1176 and 702 bp, andα-amylase and were non-tumorigenic. These amylase-producing cells can be a useful model to study the mechanisms of regulation of growth and differentiation in these cells.     

Inhibition of H-thymidine incorporation by hydroxyurea: Atypical response of mycoplasma-infected cells in culture

Contamination with Mycoplasma hyorhinis was demonstrated in long-term cultures of HeLa, BICR/M1R_K rat mammary tumor, and NV1C rat neurinoma cells, by microbiological, equilibrium sedimentation, and autoradiographic techniques. In non-infected DNA-synthesizing cells, hydroxyurea (HU) in concentrations 10⁻⁴ M typically inhibits ³H-thymidine (³H-TdR) incorporation into acid-insoluble material. This effect was lacking in the contaminated cell lines, although HU did block nuclear DNA replication, as shown by pulse-cytophotometric analyses. The response to HU could be restored to normal by supplementing the culture medium either with the anti-mycoplasma agent Tylosin or with fresh rat serum. The total ³H-activity in non-infected (or anti-mycoplasma treated) versus infected cells, in the absence of HU, was up to four times higher in the former. The data indicate that (i) incorporation of ³H-TdR into the nuclear DNA of contaminated cells was strongly reduced, probably due to a ‘scavenger effect’ (i.e. utilisation and rapid cleavage) by the mycoplasma; (ii) mycoplasmal ³H-TdR incorporation, contrary to nuclear DNA replication, was insensitive to HU in concentrations 10⁻² M. If equally valid for other species of mycoplasma, the observed phenomenon provides a criterion (together with the possibility of a rapid test) for the presence of mycoplasmal contamination in cell cultures.     

Mixotrophic growth ofChlorella sorokiniana in outdoor enclosed photobioreactor

Chlorella sorokiniana was cultured in heterotrophic or mixotrophic mode in outdoor enclosed tubular photobioreactor. The culture temperature was maintained at 32–35 °C. At night, theChlorella culture grew heterotrophically, and 0.1 M glucose was completely consumed. The biomass growth yield of glucose was 0.35 ± 0.001 g-biomass g-glucose^–1. During the day, the algal culture grew mixotrophically and the biomass growth yield was 0.49 g-biomass g-glucose^–1 in low density culture (initial biomass concentration, X_o = 2 g l^–1), 0.56 g-biomass g-glucose^–1 in medium density culture (X_o = 4 g l^–1) and 0.46 g-biomass g-glucose^–1 in high density culture (X_o = 7 g l^–1). The daily area productivity of the culture, with X_o = 4 g l^–1 corresponded to 127 g-biomass m^–2 d^–1 during the day and 79 g-biomass m^–2 d^–1 during the night. In all the cultures, the dissolved O₂ concentration increased in the morning, reached the maximum value at noon, and then decreased in the afternoon. The dissolved CO₂ concentration remained at 3 mBar in the morning and increased in the afternoon. Glycolate was not found to accumulate in culture medium.     

Aerosols as a Source of Widespread Mycoplasma Contamination of Tissue Cultures

Mycoplasma isolates were cultured from 15 antibiotic-free cell cultures obtained from a single laboratory. Complement-fixation tests showed that these isolates were antigenically related to each other but were unrelated to M. hominis type 1, M. hominis type 2, M. arthritidis, M. laidlawii type B, Mycoplasma sp. H.Ep. #2 (Barile), or M. salivarium. Examination of serum used to feed the infected cell lines revealed no Mycoplasma. Infection resulting from cross-contamination by a single Mycoplasma strain from one cell culture to another was investigated. Although the organisms were not found in the air over the work area, aerosols containing these contaminants were produced in tissue culture bottles during the trypsinization of cell monolayers. The minimal infectious dose of Mycoplasma for tissue cultures was measured, and it was determined that one organism was capable of initiating an infection in a tissue culture. The pattern of contamination and the small dose required for infection indicated that Mycoplasma contamination was spread from one tissue culture to another via aerosols. It was demonstrated that Mycoplasma can be transferred from one cell culture to another through the use of a common burette for dispensing medium.     

Establishment of Orthosiphon stamineus Cell Suspension Culture for Cell Growth

Callus of Orthosiphon stamineus could be induced successfully from petiole, leaf and stem tissues but not roots when cultured on MS medium containing different concentration of NAA (0–4.0 mg l^–1) and 2,4-D (0–2.0 mg l^–1). Highest fresh weight callus production was obtained from leaf explants and those with best friability were obtained on MS medium plus 1.0 mg l^–1 2,4-D plus 1.0 mg l^–1 NAA. Cell suspension cultures were established from these cultures. The appropriate cell inoculum size for the best cell growth was 0.75 g of cells in 20 ml culture medium. Cell suspension culture using MS medium supplemented with 1.0 mg l^–1 2,4-D promoted the best cell growth with maximum biomass of 8.609 g fresh weight and 0.309 g dry weight 24 days after inoculation. Cells that grew in MS medium supplemented with 1.0 mg l^–1 2,4-D reached the stationary growth phase in 15 days as compared to the cells that grew in MS medium supplemented with 1.0 mg l^–1 2,4-D + 1.0 mg l^–1 NAA reached the stationary phase in 24 days. MS medium supplemented with 1.0 mg l^–1 2,4-D was considered as the maintenance medium for maintaining the optimum cell growth of O. stamineus in the cell suspension cultures with 2-week interval subculture.     

Development of suspension culture of Podophyllum hexandrum for production of podophyllotoxin

A suspension culture of Podophyllum hexandrum was established. As the cultures grew, reduction in cell viability, biomass and product yield were associated with browning of culture medium, clumping of cells and drop in medium pH. Supplementation of the medium with both polyvinylpyrrollidone (PVP) and pectinase eliminated these problems. PVP at 10 g l^–1 was optimum for both growth of and product formation in P. hexandrum suspension cultures.     

Growth, metabolism and baculovirus production in suspension cultures of an Anticarsia gemmatalis cell line

The UFL-AG-286 cell line, established from embryonic tissue of the lepidopteran insect Anticarsia gemmatalis, has been identified as a good candidate to be used as a cellular substrate in the development of a process for in vitro production of the Anticarsia gemmatalis multicapsid nucleopolyhedrovirus, a baculovirus widely used as bioinsecticide. In order to characterize the technological properties of this cell line and evaluate its feasibility to use it for the large-scale production of Anticarsia gemmatalis multicapsid nucleopolyhedrovirus, UFL-AG-286 cells were adapted to grow as agitated suspension cultures in spinner-flasks. Batch suspension cultures of adapted cells in serum-supplemented TC-100 medium grew with a doubling time of about 29 h and reached a maximum cell density higher than 3.5 × 10⁶ viable cells ml⁻¹. At the end of the growth period glucose was completely depleted from the culture medium, but l-lactate was not produced. Amino acids, with the exception of glutamine, were only negligibly consumed or produced. In contrast to other insect cell lines, UFL-AG-286 cells appeared to be unable to synthesize alanine as a metabolic way to dispose the by-product ammonia. The synchronous infection of suspension cultures with Anticarsia gemmatalis multicapsid nucleopolyhedrovirus in the early to medium exponential growth phase yielded high amounts of both viral progenies per cell and reduced the specific demands of UFL-AG-286 cells for the main nutrients.     

Uracil phosphoribosyl transferase activity of mycoplasma and infected cell cultures

Summary Human H. Ep-2 and mouse 3T6 cells infected byMycoplasma hyorhinis showed an increase in [³H]uracil uptake and a more than 20-fold increase in the activity of uracil phosphoribosyltransferase (UraPRT). Uninfected cell cultures gave background levels of this enzyme activity. A survey of 16 strains of mycoplasma showed 13 to possess UraPRT activity. Rabbit kidney cells (RK13) were infected with eight different strains of four mycoplasma species known to be common cell culture contaminants. Seven of the eight cell cultures showed elevated UraPRT activities four days after infection. This enzyme activity may be of value in monitoring cell cultures for mycoplasma and aid in classification. This work was supported by Contract NO1-CP-53530 with the National Cancer Institute, National Institutes of Health, and Contract FDA 74-41 of the Food and Drug Administration, Bethesda, Maryland 20014.     

Cell Culture Derived AgMNPV Bioinsecticide: Biological Constraints and Bioprocess Issues

We have studied parameters for optimizing the Spodoptera frugiperda (Sf9) cell culture and viral infection for the production of Anticarsia gemmatalis multiple nucleopolyhedrosis virus (AgMNPV) polyhedra inclusion bodies (PIBs) in shaker-Schott or spinner bottles and bioreactors. We have assayed the k_La of the systems, initial cell seeding, cell culture volume, dissolved oxygen (DO), multiplicity of infection (MOI), nutrients consumption, and metabolites production. The medium surface oxygen transfer was shown to be higher in shaker bottles than in spinner ones, which was in direct correlation to the higher cell density obtained. Best quantitative performances of PIBs production were obtained with a SF900II medium volume/shaker-bottle volume ratio of 15% and MOI of 0.5 to 1 performed at a cell concentration at infection (CCI) of 1 to 2.5×10⁶ cells/ml in a medium containing enough glucose and glutamine. Upon infection, a decrease in the cell multiplication was observed to be dependent on the MOI used, and the μX at the exponential growth phase in infected and non-infected cultures were, respectively, of 0.2832 and 0.3914 (day⁻¹). The glucose consumption and lactate production were higher in the infected cultures (μGlucose and μLactate of, respectively, 0.0248 and 0.0089×10⁻⁸ g/cell×day in infected cultures and 0.0151 and 0.0046×10⁻⁸ g/cell×day in non infected ones). The glutamine consumption did not differ in both cultures (μGlutamine of 0.0034 and 0.0037×10⁻⁸ g/cell×day in, respectively, infected and non infected cultures). When a virus MOI of 0.1 to 1 was used for infection, a higher concentration of PIBs/ml was obtained. This was in direct correlation to a higher cell concentration present in these cultures, where a decrease in cell multiplication due to virus infection is minimized. When a MOI of 1 was used, a more effective decrease in cell multiplication was observed and a lower concentration of PIBs/ml was obtained, but with the best performance of PIBs/cell. Correlations between MOI and CCI indicate that a MOI 0.1 to 1.4 and a CCI of 10⁶ to 2×10⁶ cells/ml led to the best PIBs production performances. The virulence of PIBs produced in cultures infected at low or high MOI showed comparable DL₅₀. Culture and infection in scaling-up conditions, performed in a bioreactor, were shown to provide the cells with a better environment and be capable of potentially improving the shaker-Schott findings. For an accurate qualitative control of PIB virulence, hemolymph from AgMNPV infected Anticarsia gemmatalis was used as starting material for passages in Sf9 cells. These led to a loss of virulence among the PIBs with an increase in the DL₅₀. The loss of virulence was accompanied by a loss in budded virus titer, a decreased number of PIBs produced and an altered DNA restriction pattern, suggesting the generation of defective interference particles (DIPs). Transmission electron microscopy (TEM) studies revealed that after cell passages, PIBs lacking virions were progressively synthesized. The study described here point out the biological constraints and bioprocess issues for the preparation of AgMNPV PIBs for biological control.     

Elimination of mycoplasmas from cell cultures by a novel soft agar technique

Summary Mycoplasmal infection of cell cultures remains a significant threat to diagnostic and research procedures. In certain defined situations, curing of mycoplasmal infected cultures is a reasonable exercise. Four methods of curing were compared: treatment with BM-cycline, 5 bromouracil, use of specific antisera and treatment of infected cells suspended in soft agar with antibiotics. Antisera treatments were of low efficiency of curing: 50%. None of nine infected cell lines treated with 5-bromouracil were consistently cured of mycoplasmas. The use of BM-cycline was effective for some, but not all lines and required long periods of treatment, 12–21 days. 35 naturally or deliberately infected cultures were treated in soft agar a total of 119 times. This procedure which consisted of suspending infected cultures in soft agar containing appropriate antibiotics resulted in successful mycoplasmal elimination 118/119 times. This soft agar technique took 1–3 days. In separate studies, it was shown that certainMycoplasma fermentans strains were resisted to this and other curing methods. This may be due to their intracellular location. Such strains may be more amenable to antibiotics that penetrate mammalian cells. It is concluded that the soft agar technique is a rapid, efficient and reliable method to eliminate cell culture mycoplasmas. These studies were supported in part by grant 15748 from the National Institute of Allergy and Infectious Diseases and the W. W. Smith Charitable Trust.     

Methanol and methylamine utilization result from mutational events in Thiosphaera pantotropha

With choline as carbon source Thiosphaera pantotropha GB17 grew with a doubling time (t_d) of 6 h. The cellular yield was 55.8 g dry cell weight per mol of choline, indicating that its methyl moieties were used for growth. However, T. pantotropha was unable to grow with methanol or with methylamine as carbon source. Mutants were isolated from liquid or from solid media able to grow with methanol (Mox⁺) as carbon or methylamine as nitrogen source (Mam⁺). The Mox⁺ mutant GB17M grew with a mean t_d of 11.7h and a growth yield of 8.9 g dry cell weight per mol of methanol. Diauxic growth of strain GB17M was observed with mixtures of pyruvate and methanol as substrates in batch culture. Methanol led to the formation of methanol dehydrogenase, formate dehydrogenase, ribulosebisphosphate carboxylase and of a soluble cytochrome c-551.5. Tn5-insertional mutants defective in the thiosulfate oxidizing enzyme system or in hydrogenase acquired the Mox⁺ phenotype. However, Tn5-insertional mutants defective in either a c-type cytochrome or the molybdenum cofactor did not mutate to the Mox⁺ phenotype, indicating common functions in thiosulfate and in methanol metabolism.     

Monoclonal antibodies specific for cell culture mycoplasmas

Summary Mycoplasma infection of cell cultures is still a major problem in some laboratories. Although several methods can be used for their detection, identification is normally by serological procedures. As no commercial source for the necessary antibodies is available we have prepared monoclonal antibodies to the five mycoplasma species that account for the majority of cell culture infections. These antibodies have been characterized by the growth inhibition test (GIT), immunofluorescence, and enzyme linked immunosorbent assay (ELISA) and have shown perfect correlation in all tests when compared to conventional antisera raised in rabbits or donkeys. In addition, a monoclonal antibody toMycoplasma pneumoniae was produced.M. pneumoniae is an infrequent cell culture contaminant but is a human pathogen, and the monoclonal antibody described here could be useful in the clinical diagnosis ofM. pneumoniae infection in man. These studies were supported by Grants Al-15748 from the National Institute of Allergy and Infectious Diseases, and GM20138-07 from the National Institutes of Health, Bethesda, MD.     

Inhibition of 3H-thymidine incorporation by hydroxyurea: atpical response of mycoplasma-infected cells in culture.

Contamination with Mycoplasma hyorhinis was demonstrated in long-term cultures of HeLa, BICR/M1R_K rat mammary tumor, and NV1C rat neurinoma cells, by microbiological, equilibrium sedimentation, and autoradiographic techniques. In non-infected DNA-synthesizing cells, hydroxyurea (HU) in concentrations ? 10^?4 M typically inhibits ³H-thymidine (³H-TdR) incorporation into acid-insoluble material. This effect was lacking in the contaminated cell lines, although HU did block nuclear DNA replication, as shown by pulse-cytophotometric analyses. The response to HU could be restored to normal by supplementing the culture medium either with the anti-mycoplasma agent Tylosin or with fresh rat serum. The total ³H-activity in non-infected (or anti-mycoplasma treated) versus infected cells, in the absence of HU, was up to four times higher in the former. The data indicate that (i) incorporation of ³H-TdR into the nuclear DNA of contaminated cells was strongly reduced, probably due to a ‘scavenger effect’ (i.e. utilisation and rapid cleavage) by the mycoplasma; (ii) mycoplasmal ³H-TdR incorporation, contrary to nuclear DNA replication, was insensitive to HU in concentrations ? 10^?2 M. If equally valid for other species of mycoplasma, the observed phenomenon provides a criterion (together with the possibility of a rapid test) for the presence of mycoplasmal contamination in cell cultures.     

Photosynthetically active suspension cultures of potato spindle tuber viroid infected tomato cells as tools for studying viroid — host cell interaction

Photosynthetically active callus and cell suspension cultures were established from uninfected Lycopersicon peruvianum plants and from uninfected and potato spindle tuber viroid (PSTVd) infected plants of Lycopersicon esculentum cv. Rutgers. Viroid infection was maintained in photoheterotrophic culture on media containing 3% sucrose, but during continuous photo-mixotrophic culture in low sucrose media (1% sucrose), the level of PSTVd accumulation decreased. Photoautotrophic cell suspensions could be established with uninfected, but not with viroid infected tomato cells. As compared to uninfected cells, PSTVd infected cells grew slowly, were morphologically different in size and shape, and formed tight cell aggregates. Electronmicroscopy showed that starch accumulation in chloroplasts, deformation of the chloroplast envelope and irregular plasmalemmasomes at the cell membrane were associated with PSTVd infection.Abbreviations 2,4-D 2,4-Dichlorophenoxyacetic acid - BAP 6-benzylaminopurine - CEVd citrus exocortis viroid - CSVd chrysanthemum stunt viroid - PSTVd potato spindle tuber viroid - TMV tobacco mosaic virus - phc photoheterotrophic cell culture - mcc photomixotrophic cell culture - pcc photoautotrophic cell culture     

Growth and Plating of Cell Suspension Cultures of Datura innoxia

Suspension cultures of Datura innoxia Mill, were successfully grown on a modified Murashige and Skoog medium with 2,4–D, NAA or BAP as growth substances, provided the micronutrient levels were reduced to 1/10. Normal amounts of micronutrients were toxic. Attempts to identify the toxic elements did not succeed. Cultures grew exponentially on a shaker at 27°C in the light. Their doubling times varied from 1.1 days on 2,4–D (10^–6M) or NAA (10^?5M)+ 1 g/1 casein hydrolysate to 2.7 days on BAP (3 × 10^?7M) and 5.1 days on supraoptimal levels of 2,4-D (10^?5M). Cultures grew on NH₄⁺-N alone (from ammonium malate) or on NO₃^?-N alone. Dry weight yield was proportional to the amount of nitrate-N added (47 mg/mg N). Filtered suspension cultures containing single cells (plating cultures) could be grown in agar in petri dishes when NAA or 2,4-D were used as growth substances. Cells grew at densities above 500 units/ml in the agar. Most colonies grew from cell aggregates but division in single cells was observed. The highest plating efficiency was about 50% on 10^?6 M 2,4-D + 1 g/1 casein hydrolysate.