Mycoplasma pulmonis possesses a novel chemoattractant for B lymphocytes

Vol. 60, No. 2

INFECTION AND IMMUNITY, Feb. 1992, p. 669-674 0019-9567/92/020669-06$02.00/0 Copyright © 1992, American Society for Microbiology

Mycoplasma pulmonis Possesses a Novel Chemoattractant for B Lymphocytes SUZANNE E. ROSS,'* JERRY W. SIMECKA,1 GINGER P. GAMBILL,' JERRY K. DAVIS,2 AND GAIL H. CASSELL' 2 Departments of Microbiology' and Comparative Medicine,2 University of Alabama at Birmingham, University Station, VH 530, Birmingham, Alabama 35294 Received 20 May 1991/Accepted 21 November 1991

Mycoplasma pulmonis causes chronic murine respiratory mycoplasmosis, which is characterized by extensive peribronchial and perivascular infiltration of mononuclear cells, including B lymphocytes. B-lymphocyte recruitment into sites of inflammation is presently poorly understood but must involve directed chemotaxis of these cells in response to some external recruitment stimulus. In these studies, picogram amounts of M. pulmonis membrane protein were found to possess potent chemoattractant activity for resting rat B lymphocytes. This report is the first description of a bacterially derived chemoattractant for B lymphocytes and offers a unique opportunity to study regulation of B-lymphocyte recruitment to a site of chronic pulmonary inflammation. Furthermore, M. pulmonis membrane activation of fresh rat serum was found to produce a potent stimulus for recruitment of peritoneal and alveolar macrophages. M. pulmonis-mediated recruitment of lymphocytes and macrophages may play a significant role in the pathogenesis of murine respiratory mycoplasmosis, a role in which organisms on the bronchiolar epithelial surfaces may release proteins which can directly or indirectly promote chemotaxis of inflammatory cells from the circulation.

cells, although this mechanism has not been demonstrated in vivo. Factors described as having putative B-lymphocyte chemoattractant activity include anti-immunoglobulin (Ig) (1, 2), denatured casein (6), mast cell-derived chemotactic factor (1), lymphocyte chemotactic factor (LCF-a [11]), and interleukin-1 (1). Thus far, however, no bacterially derived chemoattractant for B lymphocytes has been found, as seen with the response of neutrophils, macrophages, T cells, and natural killer cells to bacterially derived peptides such as formyl-methionyl-leucyl-phenylalanine (12). We have examined the role of M. pulmonis, M. pulmonisactivated rat serum components, and other pathogenic rodent and human mycoplasma species in in vitro inflammatory cell recruitment, and we report here the finding of a novel chemoattractant activity for rat B lymphocytes and macrophages mediated by a surface membrane protein(s) of M. pulmonis (MPM). Furthermore, MPM-activated fresh rat serum was chemotactic for rat macrophages and neutrophils, suggesting at least one indirect mechanism of recruitment which might be mediated by a mycoplasmal protein(s).

Mycoplasmas are ubiquitous pathogens which cause chronic inflammatory disease of the respiratory tract and joints (4), possibly because of their role as potent immunomodulators of specific and nonspecific host immune responses (20). Mycoplasma pulmonis causes chronic murine respiratory mycoplasmosis (MRM) in rats (4), which is characterized by extensive pulmonary inflammation with hyperplasia of bronchus-associated lymphoid tissue and mononuclear cell infiltrates around lung airways and blood vessels. The infiltrating cells include macrophages, B and T lymphocytes, plasma cells, and non-B and non-T mononuclear cells which may be natural killer cells (7, 9). The causes of the presence and persistence of these cells in the lung in MRM are not presently understood, but several described immunomodulatory activities of M. pulmonis have been implicated. The immunomodulatory activities of M. pulmonis include activation of macrophages (8) and natural killer cells (13) and polyclonal induction of B-lymphocyte activation, proliferation, and differentiation in vivo (23) and in vitro (16, 19). M. pulmonis and other mycoplasma species are also potent inducers of class II antigen (Ta) expression on macrophages (21) and B lymphocytes (18). Such interactions with host cells may play a role in disease development and chronicity as well as autoimmune sequelae often associated with mycoplasmal infection. M. pulmonis colonizes the epithelial surfaces of the respiratory tract and is not cleared by the host immune response (5), thereby providing a continuing source of stimulation for the inflammatory reaction. However, little is known about the recruitment of B lymphocytes into sites of inflammation in MRM or other diseases. It has been generally assumed that B cells are recruited secondary to activation of host cells, such as macrophages and T lymphocytes, and that these cells subsequently release factors which may be chemotactic for B *

MATERIALS AND METHODS Animals. Ten- to 12-week-old, male, pathogen-free UAB F344 rats which were bred, raised, and maintained in Trexler-type plastic film isolators were used in all experiments. Pathogen-free status of animal colonies was monitored rou-

tinely by serological techniques (including enzyme-linked immunosorbent assay [ELISA] for serum antimycoplasmal antibodies [8]). The colony was also monitored for the presence of rat antiviral antibodies. In addition, cultural techniques were used to attempt to isolate bacterial and fungal pathogens, and retired breeders were screened by histology. The colony was negative for all pathogens, and all animals were negative for IgG and IgM anti-M. pulmonis antibodies at the time of the experiment. Chemoattractants. M. pulmonis UAB 6510, isolated and filter cloned following one rat passage of strain UAB 5782,

Corresponding author. 669

670

ROSS ET AL.

previously described (8) and used to prepare a crude membrane fraction by osmotic lysis in 200 volumes of cold distilled water after the organisms were preloaded with warm 2 M glycerol (8); lysis was followed by differential separation to first eliminate whole organisms and then recover the membrane fraction. (Membranes prepared by this method consistently show minimal contamination with cytoplasmic components as evidenced by negative staining with acridine orange for nucleic acids or cytoplasmic enzymes.) The concentration of MPM was adjusted to 200 p.g of protein per ml, the absence of viable organisms was verified by culture, and the MPM was stored in aliquots at -70°C. Mycoplasma neurolyticum type strain (ATCC 19988), Mycoplasma arthritidis (ATCC 14152), Mycoplasma pnetinoniae Eaton (type strain, ATCC 15531), and B176 (a recent clinical isolate) were grown to late log phase, washed three times in phosphate-buffered saline (PBS), and resuspended to 100 ,ug of protein per ml in RPMI 1640 medium. (Final dilution to 10 .g/ml was prepared with chemotaxis medium.) Lipopolysaccharide (LPS) (phenol extract of Escherichia coli 026:B6; Sigma), was diluted to stock concentrations in RPMI 1640 medium (GIBCO) and stored at -70°C. Control stimuli (17) were 10-5 M 1,2-dioctanoyl-rac-glycerol (Sigma), 10-5 M phorbol myristate acetate (Sigma), and 1:10 MPM- or LPS-activated rat serum (fresh rat serum incubated with 100 jig of LPS per ml or 10 jig of MPM per ml for 60 min at 37°C, heat inactivated at 56°C for 30 min, clarified by centrifugation, and filter sterilized before being frozen in aliquots at -70°C). The final dilutions of all attractants were prepared on the day of use in RPMI 1640 culture medium containing 10 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES; Research Organics, Cleveland, Ohio), 0.2% NaHCO3, 2 mM L-glutamine (GIBCO), and 0.2% bovine serum albumin (BSA). Cell collection. Rats were anesthetized by intramuscular injection of 0.25 ml of a combination of ketamine hydrochloride (Bristol Laboratories, Syracuse, N.Y.) and 10 mg of Rompun (xylazine hydrochloride; Haver-Lockhart, Shawnee, Kans.) per ml. Blood was collected by cardiac puncture, and serum was retained for ELISA for antibodies to M. pulmonis and M. arthritidis. After euthanasia of animals by cervical dislocation, spleens were aseptically removed, minced, and pressed through a 250-,um-pore-size Nitex mesh screen in cold RPMI 1640 medium with 10 mM HEPES, pH 7.2. Mononuclear cell preparations were prepared on FicollHypaque gradients (Pharmacia) as described previously (8) and counted on a Coulter Counter model Fn (Coulter Elecwas grown as

tronics, Hialeah, Fla.). Sephadex G10 column depletion of adherent cells. Mononuclear cells were sequentially passaged through two Sephadex G10 (Pharmacia Fine Chemicals, Piscataway, N.J.) columns at 37°C to remove macrophages and dendritic cells (14). This cell preparation was responsive to both B- and T-cell mitogens and contained approximately 50% each OX-19-positive (T) and surface Ig-positive (B) cells with less than 0.2% nonspecific-esterase-positive cells. Preparation of positively selected B lymphocytes. B lymphocytes were prepared by direct panning on rabbit anti-rat Ig F(ab')2-coated 100-mm-diameter petri plates as described previously (24). Briefly, plates were coated with antibody at 25

ptg/ml in Tris buffer (pH 9.5) for 1 h at room temperature.

Following five PBS washes, 3 ml of adherent cell-depleted cell suspension was added to each plate and incubated for 1 h at 4°C. Following five washes with PBS-1% fetal calf serum (FCS), adherent B cells were removed by vigorous pipeting with warm RPMI 1640 medium.

INFECT. IMMUN.

Preparation of T lymphocytes. Non-B lymphocytes from the direct panning plates were passed over nylon wool (8) to eliminate residual B lymphocytes and macrophages. Sterile, washed, nylon wool columns were equilibrated with warm RPMI 1640 medium-2% FCS. Cells were incubated on columns for 1 h at 37°C prior to elution with two column volumes of warm RPMI 1640 medium-2% FCS. Preparation of negatively selected B lymphocytes. Negatively selected B lymphocytes were prepared by indirect panning (24) of adherent cell-depleted lymphocytes on rat anti-mouse F(ab')2-coated petri dishes. Depletion of T and natural killer cells was achieved by indirect panning with a cocktail of anti-rat T and natural killer cell monoclonal antibodies including OX-19 (pan-T), W3/13 (pan-T), OX-8 (T suppressor, cytotoxic, natural killer), and W3/25 (T helper) (23), graciously provided by A. Williams and N. Barclay (Medical Research Council Immunobiology Unit, Oxford, United Kingdom). Lymphocytes (107) were incubated for 30 min at 4°C with 100 ,ul of each supernatant containing monoclonal antibodies and washed twice. Plates were prepared by adding 10 ml of 0.05 M Tris (pH 9.5) containing 25 p.g of rat anti-mouse F(ab')2 antibody per ml, incubating at room temperature for 45 min, and washing four times with sterile PBS and once with PBS with 1% FCS. Cells to be panned were adjusted to a density of 4 x 106/ml in PBS with 5% FCS and 4 ml added to each plate. Plates were incubated at 4°C for 60 min, and then nonadherent cells were transferred to a second set of rat anti-mouse F(ab')2-coated petri dishes to further deplete T cells. Cells thus prepared were over 98% surface Ig positive; less than 2% positive with OX-19, W3/25, OX-8, or nonspecific esterase; and more than 92% viable. Cells were washed and adjusted to the desired concentration(s) in culture medium containing HEPES, glutamine, gentamicin, 0.2% NaHCO3, and 5% rat serum. Preparation of macrophage populations. Peritoneal lavage cells were collected from anesthetized rats by repeated washing of the peritoneal cavity with a total of 50 ml of cold PBS with 10 U of preservative-free sodium heparin per ml. Alveolar cells were collected by tracheal lavage with 50 ml of cold PBS. Cells were washed once in RPMI 1640 medium and resuspended to 2 x 106/ml in chemotaxis medium prior to assay. Characterization of cell populations. Cell populations were characterized for viability by using trypan blue exclusion, by the presence of nonspecific-esterase-positive cells (8) and by morphological appearance on Wright-Giemsa stain. Cells were incubated with polyvalent fluorescein isothiocyanateconjugated goat anti-rat Ig (Pelfreez Biologicals) or fluorescein isothiocyanate-conjugated sheep anti-rat Ig subclass antibodies (Bioproducts for Science, Inc., Indianapolis, Ind.) for 30 min at 4°C, washed twice, and examined with a FACStar analyzer (Becton-Dickinson, Mountain View, Calif.) for cells positive for surface Ig. Cells were further characterized by indirect immunofluorescence for T (OX19), T-helper (W3/25), and T-suppressor (OX-8) cells by incubation with 20 pL1 of monoclonal antibody followed by staining with 20 R1 of fluorescein isothiocyanate-conjugated rat anti-mouse Ig F(ab')2 (Jackson Immunoresearch Laboratories, West Grove, Pa.) prior to analysis of 10,000 cells with a FACStar analyzer (8). Chemotaxis assay. The chemotactic response of cell populations was assessed by using a 48-well microchemotaxis assembly (Neuroprobe, Cabin John, Md.). For lymphocytes, autoclaved 8-p.m-pore-size polyvinylpyrrolidine-free polycarbonate filters were coated with 1 p.g of type IV mouse collagen per cm2 to enhance adhesion (17). Six replicates of

CHEMOATTRACTANT ACTIVITY OF M. PULMONIS

VOL. 60, 1992 U.

671

50

5

2 20,

0

10 C.) 6-

U

0

=L

5

o o 0

1 0.5 0.1 0.05

.0

E

0.

z

0.01 0.001 0

c

S

MM

Adh Depl

B Cell

NB

Per Macs Alv Macs

Population

FIG. 1. F344 rat mononuclear cell (MN), adherent cell-depleted (Adh Depl), B lymphocyte (B), non-B lymphocyte (NB), peritoneal macrophages (Per Macs), and alveolar macrophages (Alv Macs) were examined for chemotactic response to medium (CT), viable M. pulmonis (MP), and MPM. Asterisks indicate a response significantly different from that of the control. Peritoneal and alveolar macrophages responded primarily to viable M. pulmonis, while positively selected B lymphocytes responded best to MPM. Results in all experiments are expressed as the means of quadruplicate determinations of the number of cells per 10 oil immersion fields (OIF) (magnification, x 1,000) for three experiments (total n = 12).

each stimulus were placed in bottom wells, cells were placed in upper wells (2 x 105 in 50 RI1 for lymphocytes and S x 104 for macrophages), and chambers were incubated at 37°C in humidified 5% CO2 (3 1/2 h for lymphocytes, 4 h for peritoneal macrophages, and 6 h for alveolar macrophages). Filters were methanol fixed and Giemsa stained for analysis. Replicate chemotaxis data from multiple experiments were pooled and are expressed as the mean number of cells per 10 oil immersion fields +/- the standard deviation. Significant differences were determined by analysis of variance and Duncan's means ratio test. Statistical significance was accepted at P values of 90% by trypan blue dye exclusion, and macrophage numbers were determined by staining with nonspecific esterase (8). B cells were >97% +/- 2% MAR 18.5 positive;

Sol MPM (10) MPM (10)

-H

Ci 0

DISCUSSION The large numbers of B lymphocytes seen in the peribronchial and perivascular lesions of MRM in rats are unusual for a bacterial infection. We and other investigators (8, 9, 15, 16, 18, 19) have postulated that the ability of M. pulmonis to directly and polyclonally activate and cause proliferation and differentiation of B lymphocytes may play a role in the

50

100

150

200

250

Mean number of cells/OlF

FIG. 5. Solubilized M. pulmonis was prepared by lysis of organisms with 0.5% (vol/vol) n-octyl-p-D-glucopyranoside in the presence of 2 x 1i-' M phenylmethylsulfonyl fluoride in 0.5 M phosphate buffer (pH 7.5), followed by extensive dialysis and ultrafiltration. Solubilized MPM exhibited significant chemotactic activity at picogram concentrations (P < 0.001). OIF, oil immersion fields.

VOL. 60, 1992

CHEMOATTRACTANT ACTIVITY OF M. PULMONIS

673

TABLE 1. Effect of heat treatment and serum activation on MPM-mediated chemotaxis of rat mononuclear cell populations Mean no. of cells/10 OIF (SD)"

Stimulus (concn)

Control (0) MPM (10 ,ug/ml) Di-C-8c LPS-activated serum (l:lO)d MPM-activated serum

(1:l0)d

HIA/MPM serum (l:lO)d HIA serum (l:lO)d HIA MPM (10 p.g/ml)d

61 195 80 127

B

G10-

MN

96 174 179 76

(27)

62b (6) (16)"

(29) (12)"

50 336 82 71

(65)b (24)

(8)

(98)b (18) (26)

NB

40 47 280 24

(7) (19) (59)" (8)

Per M0

Alv M0

185 (38) 117 (18) ND 708 (33)b

7 (1) 23 (8) ND 155 (8)b

149 (62)b

19 (19)

41 (19)

27 (6)

860 (73)"

134 (8)b

39 (20) 19 (2) 201 (30)b

22 (3) 41 (10) 201 (83)"

64 (20) 47 (22) 255 (145)b

28 (16) 38 (10) 44 (15)

113 (68) 124 (29) ND

32 (11) 18 (2) ND

a Cells, prepared from F344 rat spleens as described, included the following: Ficoll-Hypaque mononuclear cells (MN), adherent cell-depleted populations

(G1-), positively selected B cells (B), nylon-wool passaged cells (NB), and peritoneal (Per M0) and alveolar (Alv M0) lavage cells. Incubation times for chemotaxis were 3.5 h for MN, G10-, B, and NB; 4 h for Per M0; and 6 h for Alv M0. Other abbreviations: OIF, oil immersion fields; ND, not done. b Results are significantly different from those of the controls (P < 0.05). c Control stimulus 1,2-dioctanoyl-rac-glycerol (Di-C-8) at 10' M.

d Fresh clarified rat serum was activated with LPS or MPM as described and then heat inactivated (HIA). Controls were serum heat inactivated prior to MPM treatment, heat-inactivated serum and MPM.

treatment of MPM results in decreased in vivo infiltration of

cells (15) as well as elimination of mitogenic (23), differentiative (19), and chemotactic activity of the organism for B lymphocytes in vitro or in vivo. Solubilization and heat treatment failed to inhibit chemoattractant activity of MPM, while they completely abrogated mitogenic activity. Therefore, if mitogenic and chemoattractant activities of the mycoplasmas are mediated by the same surface component(s), an intact surface complex or specific membrane orientation may be required for mitogenic stimulation of B lymphocytes to proliferate. Chemotaxis, however, may be mediated by a small component of the surface molecule(s) analogous to bacterial formylated peptides which are potent chemoattractants for neutrophils and monocytes (13). However, it is also possible that the two activities may be mediated by completely separate membrane components analogous again to the mitogenic lipid A and the chemotactic formylated peptides of gram-negative bacteria. The decrease in chemotactic activity observed at MPM concentrations above 20 ,ug/ml is most likely due to defective gradient formation by the chemoattractant at high concentration and subsequent loss of directed motility, as cell viability was not decreased in B lymphocytes exposed to this concentration over the incubation time of the assay. An excess of any chemoattractant stimulus will yield similar results at high concentration because of receptor saturation and subsequent failure of the cell to polarize in the absence

of a chemotactic gradient. The similar responsiveness of the positively and negatively selected B cells puts to rest concerns that preactivation of the B cells (with possible polarization or paralysis) might occur as a result of reacting cells with anti-Ig. The somewhat increased cell migration in wells with the highest concentrations of MPM (5 and 10 ,ug/ml) in the absence of a concentration gradient (same concentration of stimulus above and below the filter) in the checkerboard assay indicates that some portion of the B-lymphocyte motility observed may also be partly chemokinetic (nondirectional) in nature (22). The failure of rat lymphocytes to respond in vitro to the other pathogenic murine (and mitogenic) mycoplasma species M. arthritidis and M. neurolyticum is of interest, since lesions involving these organisms in the joint and brain also contain accumulations of mononuclear cells. It is possible that lower-passage strains of these organisms, membranes, or subcomponents might be more effective. Also, these species may depend more on indirect stimulation of chemotactic factor release by damaged host cells. Our interest at the present time is centered on mechanisms of recruitment to the lung, but it is hoped that purification of the chemotactic factor of M. pulmonis, which is under way, will yield insights

TABLE 3. Chemoattractant properties of M. pneumoniae for human peripheral blood Stimulus (concn)

TABLE 2. Chemoattractant properties of other mitogenic Mycoplasma species for rat lymphocytes Stimulus and

concn

(jig/ml) Control (0) MPM (10) M. neurolyticum (10)' M. arthritidis (10)C

Mean no. of cells/10 OIF (SD)" MN

B

59 (8) 226 (25)b 58 (2)

33 (1) 190 (31)" 49 (9)

85 (20)

42 (2)

NB

63 60 60 51

(38) (7) (8) (19)

Abbreviations are same as in Table 1. bResult is significantly different from that of the control (P < 0.05). 1 M. neurolyticum and M. arthritidis (American Type Culture Collection) were grown to late log phase, washed three times in PBS, and resuspended to 100 ,ug of protein per ml in RPMI 1640 medium. Final working dilution (1:10) was prepared in lymphocyte chemotaxis medium.

lymphocytesa

Mean no. of cells/10 OIF (SD)

Control (0) ................. 59 (21) 558 (77)b Di-C-8 (50 ng/ml) ................. 61 (30) M. pneumoniae Eaton (10 ,ug/ml)' ................. M. pneumoniae Eaton-membrane (10 ,ug/ml) ............. 94 (40) 121 (8) .......... M. pneumoniae B176 (10 ,ug/ml)'....... M. pneumoniae B176 membrane (10 ,ug/ml)'.............. 108 (34)

" Abbreviations are same as in Table 1.

Results are significantly different from those of the control (P < 0.05). ' M. pneuimoniae strains Eaton (reference strain) and B176 (a recent clinical isolate) were grown to late log phase in SP4 broth, washed three times in PBS, and adjusted to 100 ,ug of protein per ml in RPMI 1640 medium. " M. pneumoniae membranes were prepared by osmotic lysis as described. Membranes were adjusted to 100 ,ug of protein per ml in RPMI 1640 medium. Working 1:10 dilutions were prepared with lymphocyte chemotaxis medium. b

674

INFECT. IMMUN.

ROSS ET AL.

into mechanisms which may be operative in other mycoplasmal diseases. In summary, M. plulmonis is a unique and potent mediator of murine B-lymphocyte chemotaxis. The chemoattractant activity of membrane proteins of this organism for B lymphocytes and macrophages may play a significant role in their initial recruitment to the lung in the early stages of MRM, while the mitogenic activity of the organism for B cells, its ability to enhance Ia expression, and modulation of other host responses may promote in situ production of lymphocytes and macrophages to perpetuate chronic pulmonary inflammation. Further studies will be needed to identify the specific chemoattractant molecule(s) for B lymphocytes on the organism as well as to assess its importance as a factor in MRM pathogenesis and the resulting chronic inflammatory response.

8.

9.

10.

11.

12. 13.

ACKNOWLEDGMENTS

We thank Anne Pilaro for invaluable guidance with the chemotaxis assay, Phylis Otwell for technical assistance, and J. Clarke McIntosh for helpful comments. This work was partially supported by Public Health Service grants HL19741 from the National Institutes of Health and training grant 5T32 HL07553 (both to G.H.C.). S.E.R. is a Parker B. Francis Fellow in Pulmonary Research.

14. 15.

16. 1.

2.

3. 4.

5.

6.

7.

REFERENCES Berman, J. S., D. J. Beer, A. C. Theodore, H. Kornfeld, J. Bernardo, and D. M. Center. 1990. Lymphocyte recruitment to the lung. Am. Rev. Respir. Dis. 142:238-257. Berman, J. S., W. W. Cruikshank, D. J. Beer, H. Kornfeld, J. Bernardo, A. C. Theodore, and D. M. Center. 1988. Lymphocyte motility and lymphocyte chemoattractant factors. Immunol. Invest. 17:625-677. Biberfeld, G. 1977. Mycoplasma pneumoniae is a human polyclonal B cell activator. Scand. J. Immunol. 6:1145-1155. Cassell, G. H., J. R. Lindsey, H. J. Baker, and J. K. Davis. 1980. Mycoplasmal and rickettsial diseases, p. 244-259. In H. J. Baker et al. (ed.), The laboratory rat, vol. 1. Academic Press, Inc., New York. Cole, B. C., D. R. Kartchner, and D. J. Wells. 1990. Stimulation of mouse lymphocytes by a mitogen derived from Mycoplasma arthriditis (MAM). VIII. Selective activation of T cells expressing distinct V-beta T cell receptors from various strains of mice by the "superantigen" MAM. J. Immunol. 144:425-431. Czinn, S. J., and M. E. Lamm. 1989. Selective chemotaxis of subsets of B lymphocytes from gut-associated lymphoid tissue and its implications for the recruitment of mucosal plasma cells. J. Immunol. 136:3607-3611. Davis, J. K., P. A. Maddox, R. B. Thorp, and G. H. Cassell. 1980. Immunofluorescent characterization of lymphocytes in

17.

18.

lungs of rats infected with Mvcoplasma pulmonis. Infect. Immun. 27:255-259. Davis, J. K., J. W. Simecka, J. S. P. Williamson, S. E. Ross, M. M. Juliana, R. B. Thorpe, and G. H. Cassell. 1985. Nonspecific lymphocyte responses in F344 and LEW rats: susceptibility to murine respiratory mycoplasmosis and examination of cellular basis for strain differences. Infect. Immun. 49:152-158. Davis, J. K., R. B. Thorp, P. A. Maddox, M. B. Brown, and G. H. Cassell. 1982. Murine respiratory mycoplasmosis in F344 and LEW rats: evolution of lesions and lung lymphoid cell populations. Infect. Immun. 36:720-729. Gee, A. P. 1984. Advantages and limitations of methods for measuring cellular chemotaxis and chemokinesis. Mol. Cell. Biochem. 62:5-11. Hayashi, H., M. Honda, Y. Mibu, S. Yamamoto, and M. Hirashima. 1988. Natural mediators of leukocyte chemotaxis. Immunol. Rev. 162:140-170. Hugli, T. E. 1989. Chemotaxis. Curr. Opin. Immunol. 2:19-27. Lai, W. C., S. P. Pakes, C. Stefanu, and Y. S. Lu. 1987. Mvcoplasina pulinonis infection augments natural killer cell activity in mice. Lab. Anim. Sci. 37:299-303. Ly, I., and R. Mishell. 1974. Removal of macrophages by Sephadex GIO column depletion. J. Immunol. Methods 5:239245. Naot, Y., S. Davidson, and E. S. Lindenbaum. 1981. Mitogenicity and pathogenicity of Mycoplasma pulmonis in rats. Atypical interstitial pneumonia induced by mitogenic mycoplasmal membranes. J. Infect. Dis. 143:55-62. Naot, Y., and H. Ginsburg. 1978. Activation of B cells by mycoplasma mitogens. Immunology 34:715-720. Pilaro, A. M., T. J. Sayers, K. L. McCormick, C. W. Reynolds, and R. H. Wiltrout. 1990. An improved in vitro assay to quantitate chemotaxis of rat peripheral blood large granular lymphocytes. J. Immunol. Methods 135:213-224. Ross, S. E., J. K. Davis, and G. H. Cassell. 1990. Induction of hyper-la antigen expression on F344 rat B lymphocytes by Mvcoplasma pulmonis mitogen, p. 584-592. In G. Stanek (ed.), Recent advances in mycoplasmology. Gustav Fischer Verlag,

Stuttgart, Germany. 19. Ross, S. E., J. K. Davis, and G. H. Cassell. Submitted for

publication. 20. Ruuth, E., and F. Praz. 1989. Interactions between mycoplasmas and the immune system. Immunol. Rev. 112:133-160. 21. Stuart, P. M., G. H. Cassell, and J. G. Woodward. 1989. Induction of class II MHC antigen expression in macrophages by Mycoplasma species. J. Immunol. 142:3392-3399. 22. Wilkinson, P. C. 1990. Relation between locomotion, chemotaxis, and clustering of immune cells. Immunology 69:127-133. 23. Williamson, J. S. P., J. K. Davis, and G. H. Cassell. 1986. Polyclonal activation of rat splenic lymphocytes after in vivo administration of Mycoplasma pulmonis and its relation to in vitro response. Infect. Immun. 52:594-599. 24. Wysocki, L. J., and V. L. Sato. 1978. "Panning" for lymphocytes: a method for cell selection. Proc. Natl. Acad. Sci. USA 75:2844-2848.