1108
Killing of Cryptococcus neoformans by Human Peripheral Blood Mononuclear Cells Stimulated in Culture Stuart M. Levitz, Timothy P. Farrell, and Richard T. Maziarz
Evans Memorial Department of Clinical Research and Department of Medicine, University Hospital, Boston University Medical Center; Division of Hematology, Brigham and Womens Hospital, Harvard Medical School, Boston, Massachusetts
Both clinical and experimental studies have demonstrated that effective host defenses against infections with the encapsulated yeast Cryptococcus neoformans require an adequate cell-mediated immune (CMI) response [1], yet the specifics of how the normal immune system is activated to defend the host against challenge with the ubiquitous C. neoformans are incompletely elucidated. In murine models in vitro, activated macrophages, T cells, and natural killer (NK) cells have demonstrable activity against C. neoformans under defined conditions [2-6]. In vivo, the survivalof mice challengedwith C. neoformans has been reduced by silica treatment (which incapacitates macrophages and perhaps other effector cells) or by depletion of CD4+ T cells (but not NK cells) [7-9]. Partial protection against cryptococcosis has been seen after transfer of T cells from immunized to naive mice or transfer of normal spleen cells (but not NK-depletedspleen cells) into cyclophosphamide-treated mice [10, 11]. These results suggest that, at least in murine models, control of cryptococcosis requires several different cell types either to act as or to activate anticryptococcal effector cells. Humanculturedmonocyte-derived macrophages and freshly isolated alveolar macrophages have been shown capable of inhibiting but not killing C. neoformans [12, 13], whereas
Received 1 October 1990; revised 17 December 1990. Informed consent was obtained from all volunteers. US Department of Health and Human Services and Boston University Medical Center guidelines for human experimentation were followed. Grant support: National Institutes of Health (AI-00658, -25780), American Cancer Society (IM-519), and Leukemia Research Foundation. Reprints or correspondence: Dr. Stuart M. Levitz, Room E540, University Hospital, 88 E. Newton St., Boston, MA 02118. The Journal of Infectious Diseases 1991;163:1108-1113 © 1991 by The University of Chicago. All rights reserved. 0022-1899/91/6305-0026$01.00
killing can be effected by nonadherent population(s) of peripheral blood mononuclear cells (PBMC)includingNK cells, but only if anticryptococcal antibodyis present [14, 15]. While human neutrophils and monocytes can kill C. neoformans [16], several lines of evidence suggest that these cells do not have a primary role in host defenses againstthis fungus. First, there is no apparent defect in the ability of neutrophils and monocytes from patients with cryptococcosis to kill C. neoformans [16]. Second, cryptococcosis is rarely seen in patients with quantitative or qualitativedisorders of phagocytefunction [1]. Finally, in pathologic specimens from patientswithestablished cryptococcosis who are able to mount an inflammatory response, lymphocytes and macrophages predominate [17]. Thus, while it is assumed that the cell types involved in the eMI response are ultimately responsible for killing C. neoformans, heretofore fungicidalactivity of such cells in the absence of antibody has not been demonstrated in humans. Human lymphocytes proliferate when incubated with heatkilled C. neoformans in the presence of antigen-presenting cells, with a peak response seen after 7-9 daysof culture [18]. We hypothesized that a consequence of this proliferative response would be the generation of effector cells capable of killing C. neoformans. Therefore, we examined the anticryptococcalactivityofPBMC culturedwith or withoutheat-killed C. neoformans.
Materials and Methods Materials. All reagents used, except where noted otherwise, were obtained inthehighest quality available from Sigma Chemical (St. Louis). Medium used unless stated otherwise was RPMI 1640 (GIBCD, Grand Island, NY) supplemented with L-glutamine, penicillin, streptomycin, and 10% human heat-inactivated AB male serum. Flat-bottom 96-well polystyrene tissue culture plates (25860; Corning Glass Works, Corning, NY) were used for all assays in-
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
Since cell-mediated immunity (CMI) is criticalforhostdefenses againsttheencapsulated fungus Cryptococcus neoformans, the productionof human antifungal effector cells as a consequence of the CMI response wasinvestigated. Peripheral bloodmononuclear cells (PBMC) were stimulated in culture with killed C. neoformans. Stimulated (but not unstimulated) PBMC killed a subsequent inoculumof live encapsulated organisms, with maximal killingseenafter 3-7 days. Killing requiredthe presence of both adherent and nonadherentstimulatedPBMC and wasenhanced by anticapsular antibody. In contrast, unstimulated PBMC and the adherent fraction of stimulatedPBMC killed an isogenic acapsular strain of C. neoformans. These data suggest that the CMI response controls cryptococcosis by eliciting two populations of fungicidal cells actingsynergistically. Moreover, capsule, bythwarting the abilityof unstimulatedPBMC to kill C. neoformans, seems toobligate the hostto mountan immune response togenerate fungicidal cells.
JID 1991;163 (May)
Killing of C. neoformans by PMBC
where indicated, 20 JLI of anticapsular antibody. For wellscontaining stimulated PBMC, the samestrainofheat-killed andlive C. neoformans wasused. Wells wereincubated another2-168h in a 37°C, 5% C02 atmosphere. Numbers of colony-forming units (cfu) of C. neoformans werethendetermined by sonication, dilutions, andspread platesas described[12]. For each experiment, twosets of cell wells were includedcontaining yeast cells, medium, and pooled human serum but withoutPBMC. The first set wassonicated,diluted, and plated immediately. The number of cfu was used to calculate the inoculumof live organisms added per well. The secondset was incubatedat 37°C the sameamountof time as the experimental wells before being processed and plated; counts were used to calculate fungal growthin mediumnot containing PBMC. Anticapsular antibody did not affect growth of controls. Results wereexpressed as percentage changein the inoculumaccording to the formula [(cfu experimental/cfu inoculum) - 1] X 100.Therefore, a negative valueindicatesthat there were fewer cfu in the experimental wellsthan in the inoculumand therefore killing hadtaken place. However, becausesomeof the yeastcells mayhave grownwhile others were killed, the exact amountof killingcannot be calculated. Thus, a value of -35 % means that a minimum of 35% of the inoculum was killed. Similarly, a positive value indicates that there weremore cfu in the experimental wellsthan in the inoculum and therefore fungal growth had ensued. Detection of cell surfaceantigens. PBMCwereisolatedand culturedwithor without C. neoformans as above except in six-well plates. After collectionof nonadherent cells, the adherentcells were harvested using a rubber policeman in the presence of EDTA. Cells wereincubated withantigen-specific monoclonal antibodies, followed by fluorescein-conjugated goatanti-mouse IgG, IgM, and IgA(Cappel, Malvern, PA). The percentageof cells stainingwith each antibody was determined with a flow cytometer (FACScan; Becton Dickinson, Mountain View, CA). Monoclonal antibodies used to detect the indicatedcell surface antigens were as follow [23]: anti-T1IA, CD2 (T cells; gift of R. Lawton, Harvard Medical School); Leu 4, CD3 (T cells; Becton Dickinson); Leu 3a, Leu 3b, CD4 (helper T cells; Becton Dickinson); Leu 1, CD5 (T cells and B cell subset; Becton Dickinson); Leu2a, CD8(suppressor Tcells; Becton Dickinson); OKMlO, CD11b (C3bireceptor;gift ofP. Rao,OrthoPharmaceuticals, Raritan,NJ); TAC, CD25 (low-affinity interleukin-2 receptor; Ortho); anti-CR1, CD35 (C3b receptor; Becton Dickinson); R/Rl, CD54 (ICAM-1; gift ofD. Staunton, HarvardMedicalSchool); HNK 1, CD57 (natural killer cell marker; ATCC, Rockville, MD); LFA-3, CD58 (gift ofS. Burakoff, Harvard Medical School); WT31 (l3-chain oftheT cell receptor; T Cell Sciences, Cambridge, MA); and oTCR-1 (o-chain of the T cellreceptor;giftof M. Brenner, HarvardMedical School). Included in each run were negative controls using no monoclonal antibodies and positivecontrols with W6/32(ATCC), which reacts witha publicdeterminant on humanleukocyte antigen-A, -B, and-CO Statistics. MeansandSE of samplegroupswerecomparedusing the two-sample, two-tailed Student's t test.
Results
Proliferation ofPBMCin response to C. neoformans. In agreement withprevious studies [18], encapsulated heat-killed C. neoformans stimulated proliferation of humanPBMC, with
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
volving cell wells. Pooledhumanserumwasobtainedby combining serum from 10 healthy volunteers. Anticapsular antibody (gift of 1. Bennett, NationalInstitutes of Health, Bethesda, MD) was prepared by immunizing rabbits with serotypeD capsular polysaccharide and used at a subagglutinating concentration (1:4000 final dilution). C. neoformans. Organisms used were encapsulated, serotype D strain MP415 (also known as strain B3501) and acapsular, isogenicmutantstrainCAP67(ATCC 52817) (obtained from E. Jacobson, Medical College of Virginia, Richmond). In murine models, MP415 is virulent, whereas CAP67 is relatively avirulent [19,20]. Organisms were maintained on asparagineminimal agar medium, harvested, counted, and suspendedin RPMI 1640 [2, 12,21]. Under these conditions, >95 % of the C. neoformans were present as singlecells, viabilityof the cells as measuredby a microcolonyassayaveraged >98 %, and clumpingof organisms wasnot observed. The average capsule thickness on the encapsulated strain was 1.5 JLm [12]. C. neoformans were heat-killed by immersion in a boiling water bath for 30 min. Isolation of PBMC. Human peripheral blood was obtained by venipuncture from normalvolunteers. For each set of experiments, the samesourcewasnot used morethanonce. Bloodwasanticoagulated with 10 units/ml heparin and the PBMC purifiedby centrifugationon a ficoll-hypaque densitygradient [22]. PBMC werewashed twice in PBScontaining0.1% bovineserum albumin, counted, and suspended in medium at the desired concentration. Proliferation assays. Assays weredone essentially as described previously [18]. Briefly, PBMC (2 x lOS) and heat-killed C. neoformans (105) were incubated up to 15 days in cell wells containing 100 JLI of medium in a humidified 5% C02 environment at 37°C. PH]thymidine (0.5 JLCi; NEN, Boston) was added to each well during the last 24 h of culture. Plates were harvested with a harvester (PHD;Cambridge Technology, Watertown, MA)ontoglass filterpaper and countedby scintillation spectroscopy. All proliferation assays includedcontrols containing PBMC but no C. neoformans. These controls averaged 90% as assessed bytrypanblueexclusion. Separation of PBMC into adherent and nonadherent populations. Cultured PBMC were separatedon the basis of adherence to plastic. PBMC (2 x 106) were cultured in cell wells for 15 days as above. Wells were then agitated and the mediumcontainingthe suspended nonadherent cellstransferred to a newwell. Mediumwas replaced in the old well and the transfer procedure was repeated. Volumes in the cell wellswere kept constantby additionof medium to the old wells or removal of supernatantfrom the new wells after settling of the cells. For some experiments, the nonadherent cells were incubatedfor 2 h in the new wells and then added back to the adherent cells in the old wells, essentiallyby reversing the above procedure. Thus, thesewellsultimately contained both adherentand nonadherent cells. Antifungal activityofcultured PBMC. For each well, supernatant was removed and replacedwith 20 JLI of pooled humanserum, 50 JLI of mediumcontaining rv5 X 103 cfu of C. neoformans, and,
1109
1110
C?
100
0
.. -:) ..
80
E Q. .£
i=
.. . ..
40
..J :>
- • . ENCAP PLUS Ig
-20
_._._._._._ T
.
- ...... ACAP
-40
.......................................
ClI
Z
< J:
0
*
~
200
-k--
~
150
ClI Z
-60
< J:
0
-80 -100 0
2
3
4
5
6
7
-+- ENCAP
300
- ... ENCAP PLUS Ig
w
w
350
250
0 0
..................
III 1
*
ACAP
....
/
/1 ......... ......
/
......
/'/:/r--------"':-=--::.-'-"-,
100 50
/~/// //
0
/.
-50 0
2
4
5
6
7
TIME (days)
Figure 3. Kinetics of antifungal effects of stimulated (A) and unstimulated (B) peripheral blood mononuclear cells (PBMC) cultured for 2 h to 7 days with live Cryptococcus neoformans: encapsulated (ENCAP), encapsulated and opsonized with anticapsular antibody (ENCAP PLUS Ig), or acapsular (ACAP). Results are means ± SE of two experiments in triplicate. Antifungal activity of stimulated PBMC was significantly increased compared with that of unstimulated cells (P < .02) at 1, 3, and 7 days except against the acapsular strain at 1 day.
formans-stimulated PBMC could empower unstimulated PBMCto kill C. neoformans. Twice weekly during a IS-day culture of PBMC, the medium wastotally replacedwith supernatants derived from PBMC stimulated with heat-killed encapsulated C. neoformans. PBMC were then challenged with live encapsulated organisms for 24 h, after which antifungal activitywasdeterminedbycounting cfu. There were no significant differences in antifungal activity between untreated and supernatant-treated PBMC (141% and 125% increase in cfu compared with the inoculum, respectively). Kinetics ofkilling ofC. neoformans by PBMC. The above experiments examined antifungal activity after 24 h of incubation of cultured PBMC with live organisms. We next examined the effect of varying the incubation period (from 2 h to 7 days) on antifungal activity. Stimulated cultured PBMC killed C. neoformans, with a maximaldecrease in cfu compared with the inoculum seen after 3-7 days of incubation (figure 3A). This time-dependent increasein killingwasseen withboth encapsulated and acapsular organisms. In contrast, opposite results were obtained using unstimulated cultured PBMe. After 3 and7 days of incubation, substantial increases in cfucomparedwiththe inoculum ofencapsulated and acapsular organisms were observed (figure 3B). Populations of cultured PBMCresponsible for antifungal activity. For these experiments, PBMC werecultured with heat-killed C. neoformans for 15 days, separated into two populations on the basis of adherence to plastic, and challengedfor 24 h with live C. neoformans. Neither population alone decreased the number of encapsulated C. neoformans compared with the inoculum, although stimulated adherent cells inhibited growth nearly completely in the presence of anticapsular antibody (figure 4). Moreover, there was less fungal growth in wells containing adherent or nonadherent
120 ~
::::» ..J ::::» 0
90 60
~
30
w ClI z
0
< J: 0
*
ENCAP
_
ENCAP PLUS Ig
0
~
_
~ACAP
-30 -60 ADHERENT
NONADHERENT
BOTH
Figure 4. Antifungal activity of adherent or nonadherent cultured stimulated peripheral blood mononuclear cells (PBMC) against live Cryptococcus neofonnans: encapsulated (ENCAP), encapsulated and opsonized with anticapsular antibody (ENCAP PLUS Ig), or acapsular (ACAP). BarH = PBMC separated into adherent and nonadherent, then recombined. Results are means ± SE of three experiments in triplicate. Antifungal activity of combined populations was significant (P < .001)compared with that of either population alone (except for combined vs. adherent against acapsular strain); that of adherent and nonadherent populations was significant (P < .001)against acapsular strain. Under identical conditions except without PBMC, number of encapsulated and acapsular yeast cells increased by 218 % and 368 %, respectively.
populations than in control wellscontaining mediumalone. Increasing the numberof nonadherent cellsthreefold did not affect results (datanotshown). If thenonadherent PBMC were added back to the adherent PBMC, fungicidal activity was restored. For the acapsular strain, adherent (but not nonadherent)culturedPBMC were able to effect significant fungicidal activity.
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
TIME (days)
3
1112
Levitz et al.
Cellsurfaceantigens on PBMCafterstimulation withheatkilled C. neoformans. Finally, using a panel of monoclonal
Discussion We found that human PBMC cultured in the presence of heat-killed C. neoformans developed the capacity to kill a subsequent inoculum of live, encapsulated organisms. In contrast, killing did not occur when PBMC were cultured in the absence of stimuli. Killing was enhanced by the presence of specific anticryptococcal antibody and required both adherent and nonadherent populations of cells. Killing of C. neoformans by cultured PBMC increased over time: No killing was observed after 2 h, modest killing at 24 h, and maximal killing at 72-168 h. Our findings that cultured PBMC required antigenic stimulation to become competent to kill C. neoformans may help explain why cryptococcosis is so prevalent in patients with compromised eMI such as AIDS. PBMC from such patients show suboptimal lymphocyte proliferation in response to cryptococcal antigens [24, 25]. Thus, when the immunocompromised patient is exposed to C. neoformans, a CMI response that results in the generation of fungicidal cells may not develop, leading to fungal growth and clinically apparent cryptococcosis. In contrast, consistent with our in vitro data, PBMC from the immunocompetent individual challenged with C. neoformans should proliferate and generate killer cells capable of destroying the fungus. Both adherent and nonadherent populations of cultured PBMC were necessary for fungicidal activity. How this cooperative interaction results in killing of C. neoformans remains to be defined. Previously, we demonstrated that the adherent fraction of cultured PBMC bound but did not internalize most serum-opsonized encapsulated organisms [12, 26]. Thus, it is tempting to speculate that the adherent cells immobilize C. neoformans while recruiting nonadherent cells to do the actual killing. Alternatively, ongoing production of cytokines or other mediators from one cell type may be required to activate another cell type for fungicidal activity. However, the
inability of supernatants from fungus-stimulated PBMC to induce antifungal activity in untreated PBMC argues against this latter hypothesis. Whatever the mechanism, our data clearly show that at least two cell types are required to effect fungicidal activity in our system and may help explain why previous studies using purified cell populations failed to demonstrate killing. The adherent fraction ofPBMC consists of a nearly homogeneous population of the monocyte/macrophage lineage [27, 28]. However,the nonadherent fraction represents a more heterogeneous population, including T, B, and NK cells, all of which can be further separated into subsets based on phenotypic and functional characteristics. Consistent with data from previous investigators [18], after stimulation of PBMC with heat-killed encapsulated C. neoformans we observed an increase in the proportion of cells expressing the low-affinity interleukin-2 receptor, CD25. None of the 12 other cell surface markers studied on PBMC changed significantly after cryptococcal stimulation. In our experiments, both competence for fungicidal activity and the increased expression of CD25 occurred with PBMC cultured 15 days but not 7 days. However, future studies will be needed to characterize the population(s) of nonadherent cells contributing to fungicidal activity and determine whether specific factors such as cytokines are needed to activate such cells. While capsule is recognized as a major virulence factor of C. neoformans, the mechanisms whereby capsule helps the organism elude host defenses are incompletely defined. Previous studies have demonstrated that binding and internalization of encapsulated organisms by phagocytes is inhibited compared with that of mutant isolates lacking capsule [2, 26, 29]. Moreover, for most (but not all) phagocyte populations, encapsulated C. neoformans are inhibited or killed less readily than are acapsular organisms [2, 12, 21]. Furthermore, circulating capsular polysaccharide may have deleterious effects on the immune system by activating complement and inducing suppressor T cell networks [30, 31]. The present studies suggestthree other mechanisms whereby capsule alters how C. neoformans interacts with the immune system. First, capsule mutes the proliferative response of PBMC to C. neoformans. Compared with that to acapsular organisms, peak proliferation of PBMC in response to encapsulated fungi was greatly diminished in magnitude (figure 1). Second, capsule imparts on C. neoformans the ability to resist killing by unstimulated cultured PBMC. Such cells were able to kill acapsular but not encapsulated C. neoformans (figure 2). Finally, capsule endows C. neoformans with the capacity to resist killing by subsets of stimulated cultured PBMC. Adherent cells alone killed acapsular organisms, whereas both adherent and nonadherent stimulated PBMC were necessary to effect killing of encapsulated C. neoformans (figure 4). The presence in serum of specific anticapsular antibodies has been associated with a favorableprognosis in patients with
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
antibodies reactive with specific leukocyte cell surface antigens, a phenotypic analysis of PBMC from two volunteers after 7 and 15 days of stimulation with heat-killed encapsulated C. neoformans was done. Neither the level of expression nor the percentage of the mononuclear population of CD2, CD3, CD4, CD5, CD8, CDl1b, CD35, CD54, CD57, or CD58 antigens significantly changed in either person's cells as a consequence of cryptococcal stimulation. In addition, no change was noted in the percentage of T cell receptor ex/{3 or "1/{) T cell populations (data not shown). However, in agreement with previous data [18], cryptococcal stimulation resulted in an increased proportion ofPBMC exhibiting highintensity expression of CD25 (low-affinity interleukin-2 receptor). Mean expression ofCD25 was 16%,18% and 42% on days 0, 7, and 15 of culture with C. neoformans, respectively.
JID 1991;163 (May)
JID 1991;163 (May)
Killing of C. neoformans by PMBC
cryptococcosis [32]. Moreover, monoclonal anticapsular antibody was protective in a murine model of cryptococcosis [33]. In vitro, antibody-dependent inhibition of killing of C. neoformans has been documented for several (but not all) murine and human effector cell populations, including macrophages and NK cells [2, 12, 14, 15,34]. The antifungal effects of both unstimulated and stimulated cultured PBMC we observed were enhanced by rabbit anticapsular antibody (figures 2-4). Such studies lend a rationale for trials of anticapsular antibody in humans, should human monoclonal or polyclonal anticapsular antibodies become available.
References
15. Miller MF, Mitchell TG, Storkus WJ, Dawson JR. Human natural killer cells do not inhibit growth of Cryptococcus neoformans in the absence of antibody. Infect Immun 1990;58:639-45. 16. Diamond RD, Root RK, Bennett JE. Factors influencing killing of Cryptococcus neoformans by human leukocytes in vitro. J Infect Dis 1972;125:367-76. 17. Baker RD, Haugen RK. Tissue changes and tissue diagnosis in cryptococcosis. Am J Clin Pathol 1955;25:14-24. 18. Miller GPG, Puck J. In vitro human lymphocyte responses to Cryptococcus neoformans. Evidence for primary and secondary responses in normals and infected subjects. J Immunol 1984:133:166-72. 19. Fromtling RA, Shadomy HJ, Jacobson ES. Decreased virulence in stable, acapsular mutants of Cryptococcus neoformans. Mycopathologia 1982;79:23-9. 20. Jacobson ES, Ayers DJ, Harrell AC, Nicholas Cc. Genetic and phenotypic characterizationof capsule mutants of Cryptococcus neoformans. J Bacteriol 1982;150:1292-6. 21. Levitz SM, DiBenedetto DJ. Paradoxical role of capsule in murine bronchoalveolarmacrophage-mediated killingof Cryptococcus neoformans. J Immunol 1989;142:659-65. 22. MetcalfJA, GallinJI, NauseefWM, RootRK, eds. Laboratory manual of neutrophil function. New York: Raven Press, 1986:3-5. 23. Knapp W, Dorken B, Rieber P, Schmidt RE, Stein H, von dem Borne AEGK. CD antigens 1989. Blood 1989;74:1448-50. 24. Schimpff SC, Bennett JE. Abnormalities in cell-mediated immunity in patients with Cryptococcus neoformans infection. J Allergy Clin Immunol 1975;55:430-41. 25. Hoy JF, Lewis DE, Miller GG. Functional versus phenotypic analysis of T cells in subjects seropositive for the human immunodeficiency virus: a prospective study of in vitro responses to Cryptococcus neoformans. J Infect Dis 1988;158:1071-8. 26. Levitz SM, Tabuni A. Binding of Cryptococcus neoformans by human cultured macrophages. Requirements for multiple complement receptors and actin. J Clin Invest 1991;87:528-35. 27. Kaplan G, Gaudernack G. In vitro differentiation of human monocytes. Differences in monocyte phenotypes induced by cultivation on glass or on collagen. J Exp Med 1982;156:1101-14. 28. Freundlich B, Avdalovic N. Use of gelatin/plasma coated flasks for isolating human peripheral blood monocytes. J Immunol Methods 1983; 62:31-7. 29. Kozel TR, Mastroianni RP. Inhibition of phagocytosis by cryptococcal polysaccharide: dissociation of the attachment and ingestion phases of phagocytosis. Infect Immun 1976;14:62-7. 30. Macher AM, Bennett JE, Gadek JE, Frank MM. Complement depletion in cryptococcal sepsis. J Immunol 1978;120:1686-90. 31. Murphy JW, Mosley RL, Moorhead JW. Regulation of cell-mediated immunity in cryptococcosis. II. Characterization of first-order T suppressor cells (Tsl) and induction of second-order suppressor cell. J Immunol 1983;130:2876-81. 32. Diamond RD, Bennett JE. Prognostic factors in cryptococcal meningitis: a study in III cases. Ann Intern Med 1974;80:176-81. 33. Dromer F, Charreire J, Contrepois A, Carbon C, Yeni P. Protection of mice against experimental cryptococcosis by anti-Cryptococcus neoformans monoclonal antibody. Infect Immun 1987;55:749-52. 34. Miller GPG, Kohl S. Antibody-dependent leukocyte killing of Cryptococcus neoformans. J Immunol 1983;131:1455-9.
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
1. Diamond RD. Cryptococcus neoformans. In: Mandell GL, Douglas RG Jr, Bennett JE, eds. Principles and practice of infectious diseases. 3rd ed. New York: John Wiley & Sons, 1990:1980-9. 2. Levitz SM, DiBenedetto DJ. Differential stimulation of murine resident peritoneal cells by selectively opsonized encapsulated and acapsular Cryptococcus neoformans. Infect Immun 1988;56:2544-51. 3. Murphy JW, McDaniel DO. In vitro reactivity of natural killer (NK) cells against Cryptococcus neoformans. J ImmunoI1982;128:1577-83. 4. Flesch lEA, Schwamberger G, Kaufmann SHE. Fungicidal activity of IFN-gamma-activated macrophages: extracellular killing of Cryptococcus neoformans. J Immunol 1989;142:3219-24. 5. Granger DL, Perfect JR, Durack DT. Macrophage-mediated fungistasis in vitro: requirements for intracellular and extracellular cytotoxicity. J Immunol 1986;136:672-80. 6. Fung PYS, Murphy JW. In vitro interactions of immune lymphocytes and Cryptococcus neoformans. Infect Immun 1982;36:1128-38. 7. Monga DP. Role of macrophages in resistance of mice to experimental cryptococcosis. Infect Immun 1981;32:975-8. 8. Mody CH, Lipscomb MF, Street NE, Toews GB. Depletion of CD4+ (L3T4+) lymphocytes in vivo impairs murine host defense to Cryptococcus neoformans. J Immunol 1990;144:1472-7. 9. Lipscomb MF, Alvarellos T, ToewsGB, et al. Role of natural killer cells in resistance to Cryptococcus neoformans infections in mice. Am J Pathol 1987;128:354-61. 10. Lim TS, Murphy JW. Transfer of immunity to cryptococcosis by T-enrichedsplenic lymphocytes from Cryptococcus neoformans-sensitized mice. Infect Immun 1980;30:5-11. 11. Hidore MR, Murphy JW. Correlation of natural killer cell activity and clearance of Cryptococcus neoformans from mice after adoptivetransfer of splenic nylon wool-nonadherent cells. Infect Immun 1986;51: 547-55. 12. Levitz SM, Farrell TP. Growth inhibition of Cryptococcus neoformans by cultured human monocytes: role of the capsule, opsonins, the cultured surface, and cytokines. Infect Immun 1990;58:1201-9. 13. Weinberg PB, Becker S, Granger DL, Koren HS. Growth inhibition of Cryptococcus neoformans by human alveolar macrophages. Am Rev Respir Dis 1987;136:1242-7. 14. Diamond RD, Allison AC. Nature of the effector cells responsible for antibody-dependent cell-mediatedkilling of Cryptococcus neoformans. Infect Immun 1976;14:716-20.
1113
Killing of Cryptococcus neoformans by Human Peripheral Blood Mononuclear Cells Stimulated in Culture Stuart M. Levitz, Timothy P. Farrell, and Richard T. Maziarz
Evans Memorial Department of Clinical Research and Department of Medicine, University Hospital, Boston University Medical Center; Division of Hematology, Brigham and Womens Hospital, Harvard Medical School, Boston, Massachusetts
Both clinical and experimental studies have demonstrated that effective host defenses against infections with the encapsulated yeast Cryptococcus neoformans require an adequate cell-mediated immune (CMI) response [1], yet the specifics of how the normal immune system is activated to defend the host against challenge with the ubiquitous C. neoformans are incompletely elucidated. In murine models in vitro, activated macrophages, T cells, and natural killer (NK) cells have demonstrable activity against C. neoformans under defined conditions [2-6]. In vivo, the survivalof mice challengedwith C. neoformans has been reduced by silica treatment (which incapacitates macrophages and perhaps other effector cells) or by depletion of CD4+ T cells (but not NK cells) [7-9]. Partial protection against cryptococcosis has been seen after transfer of T cells from immunized to naive mice or transfer of normal spleen cells (but not NK-depletedspleen cells) into cyclophosphamide-treated mice [10, 11]. These results suggest that, at least in murine models, control of cryptococcosis requires several different cell types either to act as or to activate anticryptococcal effector cells. Humanculturedmonocyte-derived macrophages and freshly isolated alveolar macrophages have been shown capable of inhibiting but not killing C. neoformans [12, 13], whereas
Received 1 October 1990; revised 17 December 1990. Informed consent was obtained from all volunteers. US Department of Health and Human Services and Boston University Medical Center guidelines for human experimentation were followed. Grant support: National Institutes of Health (AI-00658, -25780), American Cancer Society (IM-519), and Leukemia Research Foundation. Reprints or correspondence: Dr. Stuart M. Levitz, Room E540, University Hospital, 88 E. Newton St., Boston, MA 02118. The Journal of Infectious Diseases 1991;163:1108-1113 © 1991 by The University of Chicago. All rights reserved. 0022-1899/91/6305-0026$01.00
killing can be effected by nonadherent population(s) of peripheral blood mononuclear cells (PBMC)includingNK cells, but only if anticryptococcal antibodyis present [14, 15]. While human neutrophils and monocytes can kill C. neoformans [16], several lines of evidence suggest that these cells do not have a primary role in host defenses againstthis fungus. First, there is no apparent defect in the ability of neutrophils and monocytes from patients with cryptococcosis to kill C. neoformans [16]. Second, cryptococcosis is rarely seen in patients with quantitative or qualitativedisorders of phagocytefunction [1]. Finally, in pathologic specimens from patientswithestablished cryptococcosis who are able to mount an inflammatory response, lymphocytes and macrophages predominate [17]. Thus, while it is assumed that the cell types involved in the eMI response are ultimately responsible for killing C. neoformans, heretofore fungicidalactivity of such cells in the absence of antibody has not been demonstrated in humans. Human lymphocytes proliferate when incubated with heatkilled C. neoformans in the presence of antigen-presenting cells, with a peak response seen after 7-9 daysof culture [18]. We hypothesized that a consequence of this proliferative response would be the generation of effector cells capable of killing C. neoformans. Therefore, we examined the anticryptococcalactivityofPBMC culturedwith or withoutheat-killed C. neoformans.
Materials and Methods Materials. All reagents used, except where noted otherwise, were obtained inthehighest quality available from Sigma Chemical (St. Louis). Medium used unless stated otherwise was RPMI 1640 (GIBCD, Grand Island, NY) supplemented with L-glutamine, penicillin, streptomycin, and 10% human heat-inactivated AB male serum. Flat-bottom 96-well polystyrene tissue culture plates (25860; Corning Glass Works, Corning, NY) were used for all assays in-
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
Since cell-mediated immunity (CMI) is criticalforhostdefenses againsttheencapsulated fungus Cryptococcus neoformans, the productionof human antifungal effector cells as a consequence of the CMI response wasinvestigated. Peripheral bloodmononuclear cells (PBMC) were stimulated in culture with killed C. neoformans. Stimulated (but not unstimulated) PBMC killed a subsequent inoculumof live encapsulated organisms, with maximal killingseenafter 3-7 days. Killing requiredthe presence of both adherent and nonadherentstimulatedPBMC and wasenhanced by anticapsular antibody. In contrast, unstimulated PBMC and the adherent fraction of stimulatedPBMC killed an isogenic acapsular strain of C. neoformans. These data suggest that the CMI response controls cryptococcosis by eliciting two populations of fungicidal cells actingsynergistically. Moreover, capsule, bythwarting the abilityof unstimulatedPBMC to kill C. neoformans, seems toobligate the hostto mountan immune response togenerate fungicidal cells.
JID 1991;163 (May)
Killing of C. neoformans by PMBC
where indicated, 20 JLI of anticapsular antibody. For wellscontaining stimulated PBMC, the samestrainofheat-killed andlive C. neoformans wasused. Wells wereincubated another2-168h in a 37°C, 5% C02 atmosphere. Numbers of colony-forming units (cfu) of C. neoformans werethendetermined by sonication, dilutions, andspread platesas described[12]. For each experiment, twosets of cell wells were includedcontaining yeast cells, medium, and pooled human serum but withoutPBMC. The first set wassonicated,diluted, and plated immediately. The number of cfu was used to calculate the inoculumof live organisms added per well. The secondset was incubatedat 37°C the sameamountof time as the experimental wells before being processed and plated; counts were used to calculate fungal growthin mediumnot containing PBMC. Anticapsular antibody did not affect growth of controls. Results wereexpressed as percentage changein the inoculumaccording to the formula [(cfu experimental/cfu inoculum) - 1] X 100.Therefore, a negative valueindicatesthat there were fewer cfu in the experimental wellsthan in the inoculumand therefore killing hadtaken place. However, becausesomeof the yeastcells mayhave grownwhile others were killed, the exact amountof killingcannot be calculated. Thus, a value of -35 % means that a minimum of 35% of the inoculum was killed. Similarly, a positive value indicates that there weremore cfu in the experimental wellsthan in the inoculum and therefore fungal growth had ensued. Detection of cell surfaceantigens. PBMCwereisolatedand culturedwithor without C. neoformans as above except in six-well plates. After collectionof nonadherent cells, the adherentcells were harvested using a rubber policeman in the presence of EDTA. Cells wereincubated withantigen-specific monoclonal antibodies, followed by fluorescein-conjugated goatanti-mouse IgG, IgM, and IgA(Cappel, Malvern, PA). The percentageof cells stainingwith each antibody was determined with a flow cytometer (FACScan; Becton Dickinson, Mountain View, CA). Monoclonal antibodies used to detect the indicatedcell surface antigens were as follow [23]: anti-T1IA, CD2 (T cells; gift of R. Lawton, Harvard Medical School); Leu 4, CD3 (T cells; Becton Dickinson); Leu 3a, Leu 3b, CD4 (helper T cells; Becton Dickinson); Leu 1, CD5 (T cells and B cell subset; Becton Dickinson); Leu2a, CD8(suppressor Tcells; Becton Dickinson); OKMlO, CD11b (C3bireceptor;gift ofP. Rao,OrthoPharmaceuticals, Raritan,NJ); TAC, CD25 (low-affinity interleukin-2 receptor; Ortho); anti-CR1, CD35 (C3b receptor; Becton Dickinson); R/Rl, CD54 (ICAM-1; gift ofD. Staunton, HarvardMedicalSchool); HNK 1, CD57 (natural killer cell marker; ATCC, Rockville, MD); LFA-3, CD58 (gift ofS. Burakoff, Harvard Medical School); WT31 (l3-chain oftheT cell receptor; T Cell Sciences, Cambridge, MA); and oTCR-1 (o-chain of the T cellreceptor;giftof M. Brenner, HarvardMedical School). Included in each run were negative controls using no monoclonal antibodies and positivecontrols with W6/32(ATCC), which reacts witha publicdeterminant on humanleukocyte antigen-A, -B, and-CO Statistics. MeansandSE of samplegroupswerecomparedusing the two-sample, two-tailed Student's t test.
Results
Proliferation ofPBMCin response to C. neoformans. In agreement withprevious studies [18], encapsulated heat-killed C. neoformans stimulated proliferation of humanPBMC, with
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
volving cell wells. Pooledhumanserumwasobtainedby combining serum from 10 healthy volunteers. Anticapsular antibody (gift of 1. Bennett, NationalInstitutes of Health, Bethesda, MD) was prepared by immunizing rabbits with serotypeD capsular polysaccharide and used at a subagglutinating concentration (1:4000 final dilution). C. neoformans. Organisms used were encapsulated, serotype D strain MP415 (also known as strain B3501) and acapsular, isogenicmutantstrainCAP67(ATCC 52817) (obtained from E. Jacobson, Medical College of Virginia, Richmond). In murine models, MP415 is virulent, whereas CAP67 is relatively avirulent [19,20]. Organisms were maintained on asparagineminimal agar medium, harvested, counted, and suspendedin RPMI 1640 [2, 12,21]. Under these conditions, >95 % of the C. neoformans were present as singlecells, viabilityof the cells as measuredby a microcolonyassayaveraged >98 %, and clumpingof organisms wasnot observed. The average capsule thickness on the encapsulated strain was 1.5 JLm [12]. C. neoformans were heat-killed by immersion in a boiling water bath for 30 min. Isolation of PBMC. Human peripheral blood was obtained by venipuncture from normalvolunteers. For each set of experiments, the samesourcewasnot used morethanonce. Bloodwasanticoagulated with 10 units/ml heparin and the PBMC purifiedby centrifugationon a ficoll-hypaque densitygradient [22]. PBMC werewashed twice in PBScontaining0.1% bovineserum albumin, counted, and suspended in medium at the desired concentration. Proliferation assays. Assays weredone essentially as described previously [18]. Briefly, PBMC (2 x lOS) and heat-killed C. neoformans (105) were incubated up to 15 days in cell wells containing 100 JLI of medium in a humidified 5% C02 environment at 37°C. PH]thymidine (0.5 JLCi; NEN, Boston) was added to each well during the last 24 h of culture. Plates were harvested with a harvester (PHD;Cambridge Technology, Watertown, MA)ontoglass filterpaper and countedby scintillation spectroscopy. All proliferation assays includedcontrols containing PBMC but no C. neoformans. These controls averaged 90% as assessed bytrypanblueexclusion. Separation of PBMC into adherent and nonadherent populations. Cultured PBMC were separatedon the basis of adherence to plastic. PBMC (2 x 106) were cultured in cell wells for 15 days as above. Wells were then agitated and the mediumcontainingthe suspended nonadherent cellstransferred to a newwell. Mediumwas replaced in the old well and the transfer procedure was repeated. Volumes in the cell wellswere kept constantby additionof medium to the old wells or removal of supernatantfrom the new wells after settling of the cells. For some experiments, the nonadherent cells were incubatedfor 2 h in the new wells and then added back to the adherent cells in the old wells, essentiallyby reversing the above procedure. Thus, thesewellsultimately contained both adherentand nonadherent cells. Antifungal activityofcultured PBMC. For each well, supernatant was removed and replacedwith 20 JLI of pooled humanserum, 50 JLI of mediumcontaining rv5 X 103 cfu of C. neoformans, and,
1109
1110
C?
100
0
.. -:) ..
80
E Q. .£
i=
.. . ..
40
..J :>
- • . ENCAP PLUS Ig
-20
_._._._._._ T
.
- ...... ACAP
-40
.......................................
ClI
Z
< J:
0
*
~
200
-k--
~
150
ClI Z
-60
< J:
0
-80 -100 0
2
3
4
5
6
7
-+- ENCAP
300
- ... ENCAP PLUS Ig
w
w
350
250
0 0
..................
III 1
*
ACAP
....
/
/1 ......... ......
/
......
/'/:/r--------"':-=--::.-'-"-,
100 50
/~/// //
0
/.
-50 0
2
4
5
6
7
TIME (days)
Figure 3. Kinetics of antifungal effects of stimulated (A) and unstimulated (B) peripheral blood mononuclear cells (PBMC) cultured for 2 h to 7 days with live Cryptococcus neoformans: encapsulated (ENCAP), encapsulated and opsonized with anticapsular antibody (ENCAP PLUS Ig), or acapsular (ACAP). Results are means ± SE of two experiments in triplicate. Antifungal activity of stimulated PBMC was significantly increased compared with that of unstimulated cells (P < .02) at 1, 3, and 7 days except against the acapsular strain at 1 day.
formans-stimulated PBMC could empower unstimulated PBMCto kill C. neoformans. Twice weekly during a IS-day culture of PBMC, the medium wastotally replacedwith supernatants derived from PBMC stimulated with heat-killed encapsulated C. neoformans. PBMC were then challenged with live encapsulated organisms for 24 h, after which antifungal activitywasdeterminedbycounting cfu. There were no significant differences in antifungal activity between untreated and supernatant-treated PBMC (141% and 125% increase in cfu compared with the inoculum, respectively). Kinetics ofkilling ofC. neoformans by PBMC. The above experiments examined antifungal activity after 24 h of incubation of cultured PBMC with live organisms. We next examined the effect of varying the incubation period (from 2 h to 7 days) on antifungal activity. Stimulated cultured PBMC killed C. neoformans, with a maximaldecrease in cfu compared with the inoculum seen after 3-7 days of incubation (figure 3A). This time-dependent increasein killingwasseen withboth encapsulated and acapsular organisms. In contrast, opposite results were obtained using unstimulated cultured PBMe. After 3 and7 days of incubation, substantial increases in cfucomparedwiththe inoculum ofencapsulated and acapsular organisms were observed (figure 3B). Populations of cultured PBMCresponsible for antifungal activity. For these experiments, PBMC werecultured with heat-killed C. neoformans for 15 days, separated into two populations on the basis of adherence to plastic, and challengedfor 24 h with live C. neoformans. Neither population alone decreased the number of encapsulated C. neoformans compared with the inoculum, although stimulated adherent cells inhibited growth nearly completely in the presence of anticapsular antibody (figure 4). Moreover, there was less fungal growth in wells containing adherent or nonadherent
120 ~
::::» ..J ::::» 0
90 60
~
30
w ClI z
0
< J: 0
*
ENCAP
_
ENCAP PLUS Ig
0
~
_
~ACAP
-30 -60 ADHERENT
NONADHERENT
BOTH
Figure 4. Antifungal activity of adherent or nonadherent cultured stimulated peripheral blood mononuclear cells (PBMC) against live Cryptococcus neofonnans: encapsulated (ENCAP), encapsulated and opsonized with anticapsular antibody (ENCAP PLUS Ig), or acapsular (ACAP). BarH = PBMC separated into adherent and nonadherent, then recombined. Results are means ± SE of three experiments in triplicate. Antifungal activity of combined populations was significant (P < .001)compared with that of either population alone (except for combined vs. adherent against acapsular strain); that of adherent and nonadherent populations was significant (P < .001)against acapsular strain. Under identical conditions except without PBMC, number of encapsulated and acapsular yeast cells increased by 218 % and 368 %, respectively.
populations than in control wellscontaining mediumalone. Increasing the numberof nonadherent cellsthreefold did not affect results (datanotshown). If thenonadherent PBMC were added back to the adherent PBMC, fungicidal activity was restored. For the acapsular strain, adherent (but not nonadherent)culturedPBMC were able to effect significant fungicidal activity.
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
TIME (days)
3
1112
Levitz et al.
Cellsurfaceantigens on PBMCafterstimulation withheatkilled C. neoformans. Finally, using a panel of monoclonal
Discussion We found that human PBMC cultured in the presence of heat-killed C. neoformans developed the capacity to kill a subsequent inoculum of live, encapsulated organisms. In contrast, killing did not occur when PBMC were cultured in the absence of stimuli. Killing was enhanced by the presence of specific anticryptococcal antibody and required both adherent and nonadherent populations of cells. Killing of C. neoformans by cultured PBMC increased over time: No killing was observed after 2 h, modest killing at 24 h, and maximal killing at 72-168 h. Our findings that cultured PBMC required antigenic stimulation to become competent to kill C. neoformans may help explain why cryptococcosis is so prevalent in patients with compromised eMI such as AIDS. PBMC from such patients show suboptimal lymphocyte proliferation in response to cryptococcal antigens [24, 25]. Thus, when the immunocompromised patient is exposed to C. neoformans, a CMI response that results in the generation of fungicidal cells may not develop, leading to fungal growth and clinically apparent cryptococcosis. In contrast, consistent with our in vitro data, PBMC from the immunocompetent individual challenged with C. neoformans should proliferate and generate killer cells capable of destroying the fungus. Both adherent and nonadherent populations of cultured PBMC were necessary for fungicidal activity. How this cooperative interaction results in killing of C. neoformans remains to be defined. Previously, we demonstrated that the adherent fraction of cultured PBMC bound but did not internalize most serum-opsonized encapsulated organisms [12, 26]. Thus, it is tempting to speculate that the adherent cells immobilize C. neoformans while recruiting nonadherent cells to do the actual killing. Alternatively, ongoing production of cytokines or other mediators from one cell type may be required to activate another cell type for fungicidal activity. However, the
inability of supernatants from fungus-stimulated PBMC to induce antifungal activity in untreated PBMC argues against this latter hypothesis. Whatever the mechanism, our data clearly show that at least two cell types are required to effect fungicidal activity in our system and may help explain why previous studies using purified cell populations failed to demonstrate killing. The adherent fraction ofPBMC consists of a nearly homogeneous population of the monocyte/macrophage lineage [27, 28]. However,the nonadherent fraction represents a more heterogeneous population, including T, B, and NK cells, all of which can be further separated into subsets based on phenotypic and functional characteristics. Consistent with data from previous investigators [18], after stimulation of PBMC with heat-killed encapsulated C. neoformans we observed an increase in the proportion of cells expressing the low-affinity interleukin-2 receptor, CD25. None of the 12 other cell surface markers studied on PBMC changed significantly after cryptococcal stimulation. In our experiments, both competence for fungicidal activity and the increased expression of CD25 occurred with PBMC cultured 15 days but not 7 days. However, future studies will be needed to characterize the population(s) of nonadherent cells contributing to fungicidal activity and determine whether specific factors such as cytokines are needed to activate such cells. While capsule is recognized as a major virulence factor of C. neoformans, the mechanisms whereby capsule helps the organism elude host defenses are incompletely defined. Previous studies have demonstrated that binding and internalization of encapsulated organisms by phagocytes is inhibited compared with that of mutant isolates lacking capsule [2, 26, 29]. Moreover, for most (but not all) phagocyte populations, encapsulated C. neoformans are inhibited or killed less readily than are acapsular organisms [2, 12, 21]. Furthermore, circulating capsular polysaccharide may have deleterious effects on the immune system by activating complement and inducing suppressor T cell networks [30, 31]. The present studies suggestthree other mechanisms whereby capsule alters how C. neoformans interacts with the immune system. First, capsule mutes the proliferative response of PBMC to C. neoformans. Compared with that to acapsular organisms, peak proliferation of PBMC in response to encapsulated fungi was greatly diminished in magnitude (figure 1). Second, capsule imparts on C. neoformans the ability to resist killing by unstimulated cultured PBMC. Such cells were able to kill acapsular but not encapsulated C. neoformans (figure 2). Finally, capsule endows C. neoformans with the capacity to resist killing by subsets of stimulated cultured PBMC. Adherent cells alone killed acapsular organisms, whereas both adherent and nonadherent stimulated PBMC were necessary to effect killing of encapsulated C. neoformans (figure 4). The presence in serum of specific anticapsular antibodies has been associated with a favorableprognosis in patients with
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
antibodies reactive with specific leukocyte cell surface antigens, a phenotypic analysis of PBMC from two volunteers after 7 and 15 days of stimulation with heat-killed encapsulated C. neoformans was done. Neither the level of expression nor the percentage of the mononuclear population of CD2, CD3, CD4, CD5, CD8, CDl1b, CD35, CD54, CD57, or CD58 antigens significantly changed in either person's cells as a consequence of cryptococcal stimulation. In addition, no change was noted in the percentage of T cell receptor ex/{3 or "1/{) T cell populations (data not shown). However, in agreement with previous data [18], cryptococcal stimulation resulted in an increased proportion ofPBMC exhibiting highintensity expression of CD25 (low-affinity interleukin-2 receptor). Mean expression ofCD25 was 16%,18% and 42% on days 0, 7, and 15 of culture with C. neoformans, respectively.
JID 1991;163 (May)
JID 1991;163 (May)
Killing of C. neoformans by PMBC
cryptococcosis [32]. Moreover, monoclonal anticapsular antibody was protective in a murine model of cryptococcosis [33]. In vitro, antibody-dependent inhibition of killing of C. neoformans has been documented for several (but not all) murine and human effector cell populations, including macrophages and NK cells [2, 12, 14, 15,34]. The antifungal effects of both unstimulated and stimulated cultured PBMC we observed were enhanced by rabbit anticapsular antibody (figures 2-4). Such studies lend a rationale for trials of anticapsular antibody in humans, should human monoclonal or polyclonal anticapsular antibodies become available.
References
15. Miller MF, Mitchell TG, Storkus WJ, Dawson JR. Human natural killer cells do not inhibit growth of Cryptococcus neoformans in the absence of antibody. Infect Immun 1990;58:639-45. 16. Diamond RD, Root RK, Bennett JE. Factors influencing killing of Cryptococcus neoformans by human leukocytes in vitro. J Infect Dis 1972;125:367-76. 17. Baker RD, Haugen RK. Tissue changes and tissue diagnosis in cryptococcosis. Am J Clin Pathol 1955;25:14-24. 18. Miller GPG, Puck J. In vitro human lymphocyte responses to Cryptococcus neoformans. Evidence for primary and secondary responses in normals and infected subjects. J Immunol 1984:133:166-72. 19. Fromtling RA, Shadomy HJ, Jacobson ES. Decreased virulence in stable, acapsular mutants of Cryptococcus neoformans. Mycopathologia 1982;79:23-9. 20. Jacobson ES, Ayers DJ, Harrell AC, Nicholas Cc. Genetic and phenotypic characterizationof capsule mutants of Cryptococcus neoformans. J Bacteriol 1982;150:1292-6. 21. Levitz SM, DiBenedetto DJ. Paradoxical role of capsule in murine bronchoalveolarmacrophage-mediated killingof Cryptococcus neoformans. J Immunol 1989;142:659-65. 22. MetcalfJA, GallinJI, NauseefWM, RootRK, eds. Laboratory manual of neutrophil function. New York: Raven Press, 1986:3-5. 23. Knapp W, Dorken B, Rieber P, Schmidt RE, Stein H, von dem Borne AEGK. CD antigens 1989. Blood 1989;74:1448-50. 24. Schimpff SC, Bennett JE. Abnormalities in cell-mediated immunity in patients with Cryptococcus neoformans infection. J Allergy Clin Immunol 1975;55:430-41. 25. Hoy JF, Lewis DE, Miller GG. Functional versus phenotypic analysis of T cells in subjects seropositive for the human immunodeficiency virus: a prospective study of in vitro responses to Cryptococcus neoformans. J Infect Dis 1988;158:1071-8. 26. Levitz SM, Tabuni A. Binding of Cryptococcus neoformans by human cultured macrophages. Requirements for multiple complement receptors and actin. J Clin Invest 1991;87:528-35. 27. Kaplan G, Gaudernack G. In vitro differentiation of human monocytes. Differences in monocyte phenotypes induced by cultivation on glass or on collagen. J Exp Med 1982;156:1101-14. 28. Freundlich B, Avdalovic N. Use of gelatin/plasma coated flasks for isolating human peripheral blood monocytes. J Immunol Methods 1983; 62:31-7. 29. Kozel TR, Mastroianni RP. Inhibition of phagocytosis by cryptococcal polysaccharide: dissociation of the attachment and ingestion phases of phagocytosis. Infect Immun 1976;14:62-7. 30. Macher AM, Bennett JE, Gadek JE, Frank MM. Complement depletion in cryptococcal sepsis. J Immunol 1978;120:1686-90. 31. Murphy JW, Mosley RL, Moorhead JW. Regulation of cell-mediated immunity in cryptococcosis. II. Characterization of first-order T suppressor cells (Tsl) and induction of second-order suppressor cell. J Immunol 1983;130:2876-81. 32. Diamond RD, Bennett JE. Prognostic factors in cryptococcal meningitis: a study in III cases. Ann Intern Med 1974;80:176-81. 33. Dromer F, Charreire J, Contrepois A, Carbon C, Yeni P. Protection of mice against experimental cryptococcosis by anti-Cryptococcus neoformans monoclonal antibody. Infect Immun 1987;55:749-52. 34. Miller GPG, Kohl S. Antibody-dependent leukocyte killing of Cryptococcus neoformans. J Immunol 1983;131:1455-9.
Downloaded from http://jid.oxfordjournals.org/ at University of Sussex on August 28, 2015
1. Diamond RD. Cryptococcus neoformans. In: Mandell GL, Douglas RG Jr, Bennett JE, eds. Principles and practice of infectious diseases. 3rd ed. New York: John Wiley & Sons, 1990:1980-9. 2. Levitz SM, DiBenedetto DJ. Differential stimulation of murine resident peritoneal cells by selectively opsonized encapsulated and acapsular Cryptococcus neoformans. Infect Immun 1988;56:2544-51. 3. Murphy JW, McDaniel DO. In vitro reactivity of natural killer (NK) cells against Cryptococcus neoformans. J ImmunoI1982;128:1577-83. 4. Flesch lEA, Schwamberger G, Kaufmann SHE. Fungicidal activity of IFN-gamma-activated macrophages: extracellular killing of Cryptococcus neoformans. J Immunol 1989;142:3219-24. 5. Granger DL, Perfect JR, Durack DT. Macrophage-mediated fungistasis in vitro: requirements for intracellular and extracellular cytotoxicity. J Immunol 1986;136:672-80. 6. Fung PYS, Murphy JW. In vitro interactions of immune lymphocytes and Cryptococcus neoformans. Infect Immun 1982;36:1128-38. 7. Monga DP. Role of macrophages in resistance of mice to experimental cryptococcosis. Infect Immun 1981;32:975-8. 8. Mody CH, Lipscomb MF, Street NE, Toews GB. Depletion of CD4+ (L3T4+) lymphocytes in vivo impairs murine host defense to Cryptococcus neoformans. J Immunol 1990;144:1472-7. 9. Lipscomb MF, Alvarellos T, ToewsGB, et al. Role of natural killer cells in resistance to Cryptococcus neoformans infections in mice. Am J Pathol 1987;128:354-61. 10. Lim TS, Murphy JW. Transfer of immunity to cryptococcosis by T-enrichedsplenic lymphocytes from Cryptococcus neoformans-sensitized mice. Infect Immun 1980;30:5-11. 11. Hidore MR, Murphy JW. Correlation of natural killer cell activity and clearance of Cryptococcus neoformans from mice after adoptivetransfer of splenic nylon wool-nonadherent cells. Infect Immun 1986;51: 547-55. 12. Levitz SM, Farrell TP. Growth inhibition of Cryptococcus neoformans by cultured human monocytes: role of the capsule, opsonins, the cultured surface, and cytokines. Infect Immun 1990;58:1201-9. 13. Weinberg PB, Becker S, Granger DL, Koren HS. Growth inhibition of Cryptococcus neoformans by human alveolar macrophages. Am Rev Respir Dis 1987;136:1242-7. 14. Diamond RD, Allison AC. Nature of the effector cells responsible for antibody-dependent cell-mediatedkilling of Cryptococcus neoformans. Infect Immun 1976;14:716-20.
1113
