Enhance Susceptibility of Lymphocytes to Infection by Theileria parva Sporozoites

Transcription

1 INFECTION AND IMMUNITY, Apr. 1993, p /93/ $02.00/0 Copyright C) 1993, American Society for Microbiology Vol. 61, No. 4 Tick Salivary Gland Extract and Interleukin-2 Stimulation Enhance Susceptibility of Lymphocytes to Infection by Theileria parva Sporozoites MICHAEL K. SHAW,* LEWIS G. TILNEY,t AND DECLAN J. McKEEVER International Laboratory for Research on Animal Diseases, P. O. Box 30709, Nairobi, Kenya Received 28 July 1992/Accepted 21 January 1993 Intracellular parasites show host cell specificity, and precise information on the range of host cells is a prerequisite for the identification of host molecules that account for the specificity and are involved in entry processes. The sporozoite stage of the tick-borne protozoan parasite Theileria parva binds to and enters bovine lymphocytes, but precise information on the susceptibility of other cell types present at the tick attachment site is unavailable. We quantitatively examined the susceptibility of cell types known to be present at the tick attachment site by a previously established in vitro assay. Apart from lymphocytes, sporozoites also bind to and enter macrophages and afferent lymph veiled cells; they do not bind to or enter fibroblasts, granulocytes, or erythrocytes. Sporozoites are not phagocytosed by the macrophages or veiled cells but enter them as they do lymphocytes. Since the tick attachment site is a region of cellular inflammation, we also examined the effects of agents known to be present in this area on lymphocyte susceptibility. Short-term preincubation of lymphocytes with tick salivary gland extract, with compounds that induce lymphocyte proliferation, or with interleukin-2 (IL-2), a cytokine produced by activated lymphocytes, increased host cell susceptibility by between 30 and 60%o. The IL-2-induced increase in host cell susceptibility could be prevented by treating the lymphocytes with the monoclonal antibody IL-A 111, which reacts with the bovine IL-2 receptor alpha chain and inhibits IL-2-driven cell proliferation. The changes induced by tick salivary gland extract and IL-2 occurred in less than 90 min. Similarly, peripheral blood mononuclear cells from an animal previously immunized with a nonrelated antigen (trypanosome variant surface glycoprotein) and stimulated in vitro with the same antigen showed increases in host cell susceptibility of between 70 and 125%. In contrast, treatment of lymphocytes with gamma interferon did not induce any increase in host cell susceptibility. For any intracellular parasite, the initial event in the entry process must involve some form of ligand-receptor interaction between components on the surfaces of both the invading organism and the host cell. The range of susceptible host cells will therefore be determined by the molecular nature of these initial interactions. Thus, if an invading organism exhibits a broad host cell specificity, it must use a range of common or closely related host surface molecules. Conversely, organisms which have a more selective host cell specificity will use a more restricted set of host surface molecules to facilitate attachment and entry into the host cells. Precise information on the range and nature of susceptible host cells is therefore a prerequisite for identification of the host cell surface molecules involved in the process of parasite entry. Ticks are important vectors of diseases of both human and veterinary significance, including Lyme disease, Rocky Mountain spotted fever, Crimean-Congo hemorrhagic fever, and a variety of other rickettsial, protozoan, and viral diseases. We are interested in Theileria parva, an intracellular protozoan parasite of cattle that is transmitted by the brown ear tick, Rhipicephalus appendiculatus. The parasite causes an acute and usually fatal disease of cattle known as East Coast fever and is a major economic constraint to livestock development in east, central, and southern Africa (25). The sporozoite stage of the parasite is inoculated during * Corresponding author. t Present address: Department of Biology, University of Pennsylvania, Philadelphia, PA tick feeding, enters lymphocytes, and differentiates into a multinucleate schizont (19, 20, 21). Kinetic studies have shown that infection first becomes patent in the lymph node that drains the site of sporozoite inoculation. Whether sporozoites reach the node by travelling freely in the lymph or in newly infected lymphocytes or other cell types remains to be established. While inoculation of sporozoites may commence soon after tick attachment, by 4 days a considerable infiltrative lesion has formed at the attachment site (32). A number of leukocyte populations are represented in this infiltrate, including polymorphonuclear leukocytes, lymphocytes, and macrophage-like histiocytes. T. parva sporozoites have been shown to invade and become established in all subpopulations of bovine lymphocytes (2, 6, 8, 16, 23, 26), but there is conflicting evidence about the susceptibility of cells of the granulocyte-macrophage lineage, which constitute the majority of cells at the attachment site, to infection and transformation (2, 10, 11, 22) İt is possible that sporozoites invade a wide range of host cells but develop in only a limited range of lymphoid cells. Precise information on the susceptibility of the bovine cells present at the tick attachment site to invasion by T. parva sporozoites is unavailable. In addition, little is known about the subpopulations of lymphocytes that are susceptible to infection and transformation by T. parva. What is needed is precise, quantitative information on the host cells that are not only susceptible to invasion but also capable of supporting the growth and differentiation of the parasite, as well as information on the nature of permissive host cells. In the present study, we have used an established in vitro 1486

2 VOL. 61, 1993 INCREASED LYMPHOCYTE SUSCEPTIBILITY TO T. PARVA 1487 infection assay (29) to evaluate the capacity of sporozoites to bind to and invade host cell populations that are found in the acute inflammatory lesion that forms around the tick attachment site (32). We report that sporozoites in vitro are long-lived and bind to and invade a limited range of host cells. However, the susceptibility of these cells to invasion can be increased dramatically by agents that are likely to be present at the tick attachment site, including tick salivary gland extract and interleukin-2 (IL-2). Thus, it appears that the tick attachment site is an advantageous environment for a sporozoite trying to enter and become established within a new host. MATERIALS AND METHODS Parasites. T. parva (Muguga stock) and the tick R. appendiculatus (Muguga stock), maintained in the Tick Unit at the International Laboratory for Research on Animal Diseases, were used in all experiments. Ticks were infected as nymphs by feeding on experimentally infected cattle, after which the engorged nymphs were maintained at 23 to 25 C and 80% relative humidity and allowed to molt to the adult stage. Sporozoite suspensions were prepared from the salivary glands of infected adult ticks that had fed on a rabbit for 4 days (3, 29). Host cell isolations. Peripheral blood lymphocytes (PBLs) were isolated from the defibrinated blood of Boran (Bos indicus) cattle as described previously (17). The lymphocytes were washed three times in Alsever's solution and resuspended in RPMI 1640 culture medium containing 10 mm HEPES (N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid) buffer (ph 7.2) and supplemented with 10% heat-inactivated fetal bovine serum with 2 mm glutamine, 50,g of gentamicin per ml, and 5 x 10-5 M 2-mercaptoethanol (RPMI complete culture medium). Bovine skin fibroblasts, isolated from skin snips, were grown as monolayers in RPMI 1640 complete culture medium. The fibroblasts were used between passages 3 and 6. Neutrophils (98% pure) were prepared from the erythrocyte layer of Ficoll-Paque (Pharmacia, Uppsala, Sweden) gradients by hypotonic lysis of the erythrocytes (2). Monocytes were isolated from bovine peripheral blood by allowing them to adhere to the surface of plasma-gelatincoated plastic tissue culture flasks as described previously (13). The cells were released with trypsin-edta, washed several times with RPMI complete culture medium, and kept in suspension by continual gentle mixing for 60 min or longer before use. Treatment of cells or sporozoites with trypsin has been reported previously not to prevent sporozoite entry or the subsequent development of the parasite (21). Lung macrophages were obtained postmortem by lung lavage into RPMI 1640 culture medium. The collected cells were filtered through gauze, and the erythrocytes and debris were removed by centrifugation through Ficoll-Paque (17). The macrophages were then resuspended in RPMI complete culture medium. Afferent lymph veiled cells were isolated from afferent lymph collected by cannulation of the efferent precapsular lymphatic following surgical removal of the node (7). The veiled cells were prepared from overnight collections of lymph as described by McKeever et al. (18) and resuspended in RPMI complete culture medium. Infection of cells with T. parva sporozoites. The different cell types used were infected in vitro by incubation with sporozoites derived from the salivary glands of infected adult ticks (3). To standardize the infection procedure, 250 to 300,u1 of sporozoite suspension (2,000 infected salivary gland acini per ml) was added to 200 pul of cell suspension (5 x 106 cells per ml) in a 1.5-ml Eppendorf tube, and the mixtures were incubated for 30 min at 37 C. To maximize sporozoite-host cell contact during the incubation period, the samples were mixed by inversion at 2- to 3-min intervals. At the end of the incubation period, the samples were fixed in suspension and processed for electron microscopy. In vitro survival of isolated sporozoites. Sporozoites were isolated in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum and incubated for between 0 and 360 min at 37 C before the addition of lymphocytes. The sporozoite-lymphocyte mixtures were incubated for 30 min at 37 C, and the samples were fixed in suspension and processed for electron microscopy. Effects of various agents implicated in activation of lymphocytes on host cell susceptibility. (i) Concanavalin A. Concanavalin A-stimulated lymphoblasts were prepared as described by Goddeeris and Morrison (12) and resuspended in RPMI complete culture medium. Mixtures of freshly isolated blast cells and sporozoites were incubated for 30 min at 37 C before being fixed in suspension and processed for electron microscopy. (ii) Tick salivary gland extract. Uninfected salivary glands from adult R. appendiculatus, which had been either unfed or fed for 4 days, were homogenized and centrifuged (1,500 x g for 5 min) to remove debris, and the resultant supernatant was saved. Freshly isolated lymphocytes obtained from cattle which had been maintained under acaricidal control were incubated with salivary gland extract (from the equivalent of 10 glands per ml) for 1 h at 37 C. After incubation, sporozoites were added, and the mixture was incubated for a further 30 min at 37 C before being fixed and processed for electron microscopy. (iii) Antigen-specific stimulation of PBMs. Peripheral blood mononuclear cells (PBMs) were collected from an adult Boran steer (B. indicus) that had been immunized with 200,ug of Trypanosoma brucei brucei variant surface glycoprotein (VSG) (ILTat 1.3) prepared as described by Cross (4). The PBMs were isolated as described previously by Goddeeris and Morrison (12). PBMs (5 x 105) were incubated in 2 ml of RPMI culture medium without HEPES and containing VSG at a final concentration of 5,ug/ml for 24 h or 4 days at 37 C in a humidified atmosphere of 5.0% CO2 in air. Mixtures of sporozoites and pretreated PBMs were incubated for 30 min at 37 C and processed for electron microscopy as described below. (iv) IL-2, gamma interferon, and phorbol ester. Freshly isolated PBLs were incubated for 1 h at 37 C in RPMI complete culture medium containing either 5 or 10 U of human recombinant IL-2 per ml, after which sporozoites were added, and the mixture was incubated for 30 min at 37 C before fixation. As a control, cooled PBLs were incubated on ice for 10 min with the monoclonal antibody IL-A 111, which is specific for bovine IL-2 receptor alpha (24), before the addition of 10 U of IL-2, and the cells were incubated at 37 C for 2 h, after which sporozoites were added, and the mixtures were incubated for a further 30 min at 37 C. PBLs were also incubated with IL-A 111 for 2 h before the addition of sporozoites. PBLs were also incubated for 1 to 2 h at 370C with 100 U of gamma interferon or 50 ng of phorbol-12-myristate-13- acetate (PMA; Sigma Chemical Co., Poole, United Kingdom) per ml, after which the sporozoite suspension was

3 1488 SHAW ET AL. added, and the mixture was incubated for 30 min at 37 C before fixation. Electron microscopy. After incubation, the various sporozoite-host cell mixtures were fixed in suspension by the addition of an equal volume of a freshly prepared solution containing either 2% glutaraldehyde (from an 8% stock supplied by Electron Microscope Sciences, Fort Washington, Pa.), 2% osmium tetroxide, and 0.05 M phosphate buffer (ph 6.2) or 2% glutaraldehyde in 0.05 M phosphate buffer (ph 7.2). With the first fixative, fixation was carried out on ice for 40 to 45 min, after which the samples were pelleted, washed several times in distilled water to remove excess phosphate, and then stained en bloc with aqueous 0.5% uranyl acetate for 6 to 16 h. Glutaraldehyde-fixed samples were fixed at room temperature (22 C) for 1 to 2 h, pelleted, postfixed in buffered 1% osmium tetroxide, washed, and stained en bloc with uranyl acetate as described above. The samples were dehydrated with either ethanol or acetone and embedded in an Epon-Araldite mixture. Ultrathin sections (50 to 70 nm thick) were collected on uncoated copper grids, double stained with aqueous uranyl acetate and lead citrate, and examined in a Zeiss EM 1OA electron microscope. Quantitation. In all experiments, an excess of sporozoites (>10 sporozoites per host cell) was added to the host cell suspensions, and thus the sporozoites would have had ample opportunity to bind to and enter susceptible cells during the 30-min incubation period. To quantify the susceptibility of the different types of potential host cells used in this study, we calculated an infection index for each individual experiment. The infection index is defined as the percentage of host cells showing either surface-bound or fully internalized sporozoite profiles in one section. To determine the infection index, we cut a thin section through each pellet of infected cells and counted the number of infected cells (that is, cells with sporozoites either bound to or fully internalized within the cells) per 300 to 500 cells included in that section. For each individual experiment, we also calculated the sporozoite internalization index, which is defined as the percentage of cell-associated sporozoites that are fully internalized within an infected cell. We have previously established (29) that this method of quantitation gives both reliable and reproducible results, even though the values obtained probably represent an underestimation of the actual or true infection levels, as cells showing no evidence of infection in one particular section may be seen to be infected on adjacent sections. Since our experiments were carried out with sporozoites from different batches of infected ticks and bovine cells obtained by a variety of different methods from several cattle, we always performed a control lymphocyte infection with each set of experiments. The infection rates for these control assays ranged from 23.4 to 33.6% (average, 28.5% + 3.6%), which is similar to the range obtained previously (29). Thus, we are confident that the relative differences in the susceptibility of the different bovine cell types used in this study reflect real differences and are not the result of variations between batches of sporozoites. For each experiment, both the infection index and the proportion of sporozoites fully internalized as a percentage of the total number of cell-associated sporozoites (i.e., bound and entered) were calculated. cm cui 30 C 0) INFECT. IMMUN Time: mins FIG. 1. Survival and infectivity of sporozoites in vitro. Isolated sporozoites were incubated in vitro at 37 C in RPMI 1640 complete culture medium for various times before the addition of lymphocytes. The mixtures of sporozoites and lymphocytes were incubated for 30 min at 37 C before fixation for electron microscopy. The infection index (shaded bars) is defined as the percentage of lymphocytes showing either surface-bound or fully internalized sporozoites in one section. Sporozoite internalization (hatched bars) is defined as the percentage of cell-associated sporozoites that are fully internalized within an infected lymphocyte. Values represent the means ± standard deviations from two separate experiments. RESULTS Sporozoites enter bovine lymphocytes rapidly (in less than 3 min at 37 C), and entry involves a defined series of events: initial binding, zippering and internalization, and escape of the parasite into the host cell cytoplasm (29). It is important to distinguish between phagocytosis, such as occurs when a macrophage engulfs an opsonized bacterium (see reference 1, p ), and sporozoite entry into lymphocytes, even though both modes of entry involve a sequential zippering mechanism. During sporozoite entry, the surface coat is lost concomitant with the zippering process, which results in the close apposition of the host and parasite surface membranes (29, 33). Furthermore, although sporozoite binding is affected by cytochalasin (29), the entry process does not appear to involve either the accumulation of actin filaments or the formation of host cell pseudopodia at the site of entry. In contrast, in phagocytosis, there is a much wider and more variable gap between the host cell and the phagocytosed particle, and entry of a bacterium into a cell results in an accumulation of actin filaments around the particle being phagocytosed (14) and is completely inhibited by cytochalasin. Isolated sporozoites remain viable and infective for up to 6 h in vitro. Since we do not yet know where infection of host cells occurs in vivo, we initially examined the in vitro survival of isolated sporozoites in terms of their ability to bind to and subsequently enter and establish themselves within lymphocytes. The proportion of sporozoites entering the host cell decreased over the experimental period. However, sporozoites were still able to bind to and enter susceptible cells even after 6 h of incubation in medium containing heat-inactivated serum (Fig. 1). Sporozoites invade a limited range of host cells. For all the experiments described below, we used a standard method of infection in which an excess of sporozoites was added to the various populations of cells and the mixtures were incubated at 37 C for a 30-min period. Thus, in a single section through the different pellets of cells, we can determine the percent infection as defined in Materials and Methods for that

4 VOL. 61, 1993 INCREASED LYMPHOCYTE SUSCEPTIBILITY TO T. PARVA ) 50- cm co.q- 40- D , E- 20- Fx, E A = Lymphocytes B - Monocytes C - Lung macrophages D - Veiled cells E - Fibroblasts / / ~~~F- Granulocytes G Erythrocytes 14KPjv 10- KSX,tE 1FX ; A B c D E F G FIG. 2. Susceptibility of various types of bovine cells present at the tick attachment site to infection by sporozoites. Mixtures of sporozoites and cells were incubated for 30 min at 37 C before fixation. The infection index (shaded bars) and sporozoite internalization rate (hatched bars) are defined in the legend to Fig. 1. Values represent the means + standard deviations from between three and seven separate experiments. specific cell type, and therefore we can compare the relative levels of infection in the different cell types. The relative levels of infection of the different cell types by sporozoites are shown in Fig. 2. Apart from PBLs, the only other cell types in which any sporozoite binding and internalization were observed were veiled cells and, to a much lesser extent, monocytes and macrophages (Fig. 2). In only a very few cases (less than 1%) did sporozoites bind to granulocytes and skin-derived fibroblasts, and sporozoites were never observed to bind to bovine erythrocytes (Fig. 2). To confirm that the failure of sporozoites to bind to or enter fibroblasts or granulocytes was not due to our isolation procedures or to nonviability of the sporozoites, we examined a mixed population of lymphocytes and granulocytes. In this case, sporozoites bound to and entered only the lymphocytes. Granulocytes and lymphocytes could be easily differentiated by the shape and structure of the nucleus and the presence of various types of granules within the granulocytes. The ability of sporozoites to invade bovine monocytes and macrophages has been described previously (10). However, while the majority of sporozoites entered macrophages as they did lymphocytes, a small proportion of sporozoites were phagocytosed by the macrophages and presumably destroyed in a phagolysosome (Fig. 3). Veiled cells differ from blood monocytes and macrophages in morphology and in their ability to process and present antigens (18). Because of the superiority of veiled cells in presenting antigens and their likely origin in the skin, we investigated the process of sporozoite entry into these cells in greater detail. The majority of invading sporozoites enter veiled cells by a process similar to that described for entry into lymphocytes (9, 29) (Fig. 4). However, while sporozoite entry into lymphocytes is not associated with any obvious accumulation of actin at the site of entry (29), entry into veiled cells is characterized by the presence of significant amounts of actin filaments within the host cell around the entering sporozoite (Fig. 4B). After internalization, sporozoites were observed in the process of discharging their rhoptries and microspheres as a prelude to the dissolution of the enclosing host cell membranes and escape of the parasite into the host cell cytoplasm (Fig. 4C). However, unlike the situation in lymphocytes, few host cell-derived microtubules were observed 6t. 'i>' FIG. 3. Part of a macrophage showing the presence of sporozoites (S) both within the cytoplasm in the process of escaping from the enclosing host cell membrane and within phagolysosomes (arrows). The sporozoites within the phagolysosomes have an abnormal appearance. to be associated with the newly escaped parasite in the veiled cells, even though the surface of the parasite has a prominent, 15- to 20-nm-thick surface coat (Fig. 4D). Although no extensive time course studies were performed, the rate of entry of individual sporozoites into veiled cells was similar to that of their entry into lymphocytes, with sporozoites already fully escaped into the veiled-cell cytoplasm within 15 min (data not shown). While sporozoites can enter and escape from the enclosing host cell membrane and subsequently develop into small, multinucleate schizonts within veiled cells (unpublished observations), we have thus far been unable to establish and maintain a transformed, infected cell line in vitro. Can the proportion of susceptible lymphocytes be altered in vitro? Since the tick attachment site is a region of cellular infiltration and inflammation (32), we examined the effects of various agents that might alter the level of host cell susceptibility. In pilot experiments, we examined the ability of sporozoites to invade concanavalin A-stimulated lymphocytes and found that there was a significant increase in the level of host cell susceptibility (Fig. 5A). In view of this result, we examined the effects of more physiologically relevant factors that might be present at the tick attachment site. First, we incubated lymphocytes with extracts from homogenized, uninfected salivary glands from both unfed and 4-day-fed adult ticks. In both cases, there ~~~~~A

5 x l: t...~~~?igj. 4. Sprzot etyinoveldcel. A owpwe irorphsoin hegnraorhlgyoaviedcllwtsoozie present inside the cell (arrows). (B) Sporozoite (S) in the process of entering a veiled cell; the sporozoite and veiled cell membranes "zipper" and establish intimate contact, with the two membranes separated by a thin layer of dense material. Note the accumulation of host cell actin filaments (*) around the sporozoite as it enters the veiled cell. (C) Once fully internalized, the sporozoite escape from the enclosing host membrane. The process of escape is preceded by separation of the apposed membranes, which occurs concomitant with discharge of the rhoptries and microspheres. Note the presence of sporozoite-derived fuzzy material (arrows) on the surface of the sporozoite. (D) Within 30 min of invasion, the parasite (P) has escaped from the enclosing host cell membrane. The freed parasite is surrounded by a layer of parasite-derived fuzzy material, but, unlike in the lymphocyte, no host-derived microtubules are present surrounding the parasite ,.Ip

6 VOL. 61, 1993 INCREASED LYMPHOCYTE SUSCEPTIBILITY TO T. PARVA 1491 A B a1) 0) Co c a) L- a) a. a1) 0) 4- CR a1) a) a C a) 0D 60- ( c 50- o) 40- a) al Control Con A Control unfed 4 day fed Salivary gland extract 0 Control 24 h 4 days FIG. 5. Effect of treating lymphocytes or PBMs in vitro with various agents implicated in cell activation on sporozoite invasion: (A) concanavalin A; (B) tick salivary gland extract (equivalent of 10 glands per ml); (C) PBMs from an animal immunized with trypanosome VSG incubated with the same antigen (5,ug/ml) for 24 h or 4 days. Host cells were treated as described in Materials and Methods and then incubated with sporozoites for 30 min at 37 C before fixation. The controls were either untreated lymphocytes or, for VSG pretreatment (panel C), unstimulated PBMs from the same immunized animal. The infection index (shaded bars) and sporozoite was an increase (30 to 50% in comparison with untreated lymphocytes) in the level of infection (Fig. 5B) which was induced within a relatively short time (i.e., less than 90 min). We also examined the susceptibility of PBMs isolated from a cow immunized with a nonrelated antigen (T. brucei brucei VSG) and stimulated in vitro with this antigen for 24 h or 4 days. There was a significant increase (between 80 and 125%) in the level of host cell susceptibility compared with unstimulated PBMs from the same animal (Fig. 5C). Furthermore, this treatment increased not only the total number of infected cells but also the number of sporozoites bound to or fully internalized within the infected cells (Fig. 6). In light of these results, we examined the effects of IL-2, gamma interferon, and the phorbol ester PMA on host cell susceptibility. IL-2, a cytokine produced by T cells, is an extracellular signal molecule involved in the activation of lymphocytes. Treatment of PBLs with IL-2 resulted in a significant increase (60 to 70%) in host cell susceptibility (Fig. 7A). This increase in host cell susceptibility was dramatically inhibited by pretreating PBLs with the monoclonal antibody IL-A 111, which is reactive with the bovine IL-2 receptor alpha chain and is a strong inhibitor of IL-2- dependent proliferation of bovine lymphocytes (24) (Fig. 7B). Similarly, treatment of PBLs with PMA, which acts on an early intracellular component of the activation cascade involving protein kinase C, also resulted in an increase (30%) in host cell susceptibility (Fig. 7C). Moreover, these increases in host cell susceptibility could also be induced within a relatively short time (i.e., less than 90 min). In contrast, treatment of PBLs with gamma interferon for up to 2 h did not induce any increase in host cell susceptibility (Fig. 8). This result, as well as the downregulation of host cell susceptibility following treatment with IL-A 111, suggests that the observed increase in host cell susceptibility following IL-2 or PMA treatment was unlikely to be a generalized, nonspecific effect. DISCUSSION Sporozoites bind to and invade a very limited population of host cells. Although the process of sporozoite entry into lymphocytes in vitro has been well documented (9, 29), the actual host cells that the parasite can invade (as opposed to those cells in which the parasite can develop and which it can transform [2]) have not been examined systematically. The results of the present study, which show that T. parva sporozoites invade only a limited range of host cells, are in agreement with previous studies, in which the parasite was found to develop only in a restricted range of bovine cells (2, 6, 8, 16, 23, 26). It should be emphasized that in our infection assay, the sporozoites are given maximum opportunity to invade the various types of bovine cells used. Moreover, in infection experiments involving mixtures of these various nonsusceptible bovine cells and lymphocytes, sporozoites preferentially invade only the lymphocytes. Thus, we are confident that sporozoites are capable of recognizing and invading only limited populations of lymphocytes, veiled cells, and, to a lesser extent, macrophages. The fact that sporozoites invade only a limited range of internalization rate (hatched bars) are defined in the legend to Fig. 1. Results represent the means + standard deviations from three separate experiments.

7 1492 SHAW ET AL. INFECT. IMMUN..-'.'. o.. I - p.. A <.1..).r.lf..r. FIG. 6. (A) Micrograph of a lymphocyte treated for 60 min with IL-2 (10 U) and subsequently incubated for 30 min with sporozoites, showing the presence of a large number of sporozoites (S) within the lymphocyte cytoplasm. Many of the internalized sporozoites either have escaped from the enclosing host cell membrane or are in the process of doing so (arrows). (B) Micrograph of a representative lymphocyte from an untreated control, showing the presence of only one sporozoite (arrow) internalized within the cell. host cell types indicates that the host cell surface molecule(s) recognized by the parasite must be selectively specific for this range of cells. This fact, together with the observation that the number of susceptible cells within a population of lymphocytes can be experimentally altered, gives us a possible means of identifying the host cell surface molecules involved in sporozoite invasion of bovine lymphocytes. Fate of sporozoites in veiled cells and macrophages. T. parva sporozoites can invade and become established within veiled cells and macrophages in a manner similar to their entry into lymphocytes. However, although the parasite seems capable of developing into a small, multinucleate schizont stage within veiled cells in vitro, the infected cells do not divide and accordingly do not establish schizontinfected, transformed cell lines (unpublished observations). Since veiled cells are involved in the endocytosis and subsequent presentation of antigens to other cells of the immune system, the question arises whether veiled cells could transfer internalized sporozoites or even young schizonts to a less-differentiated cell (e.g., a lymphocyte), which in turn is capable of cell division and blastogenesis. While the transfer of T. parva schizonts between host cells may occur at very low frequencies in vivo (20), there are several reasons why such a transfer of parasites from veiled cells to uninfected lymphocytes would appear to be unlikely or at best an extremely rare event. First, the parasite, once established within the host cell cytoplasm, no longer possesses the surface coat present on the sporozoite, a coat which is involved in the binding and entry process (29). Thus, if the parasite was freed from the first host cell, it would be unable to bind to and enter a second host cell. Second, the established intracellular parasite does not contain any of the secretory organelles (e.g., rhoptries and microspheres) necessary for escaping from the enclosing host cell membrane. Thus, even if an established parasite were freed from a veiled cell, it is extremely difficult to envisage how this freed parasite, assuming it could find and enter a lymphocyte, could escape from the enclosing host cell membrane and establish itself within the lymphocyte cytoplasm. Activation of lymphocytes by tick salivary gland extract, IL-2, and antigens enhances host cell susceptibility. Because the tick attachment site is a region of inflammation containing lymphoblasts and activated macrophages, it is possible that sporozoites preferentially invade immunologically activated cells. Accordingly, we examined the susceptibility of

8 VOL. 61, 1993 INCREASED LYMPHOCYTE SUSCEPTIBILITY TO T. PARVA 1493 A 70 OD 20 10, o 20 '/,,, 0 B 60- Control 5 IU 10 IU IL-2 PBLs pretreated 50-2 h - 370C XX m H a-20-0, C 5o a- 40 Control IL-A 111 IL-A 111 IL-A IL-2 Control 50 ng/ml PMA FIG. 7. Effect of treating lymphocytes with (A) IL-2 (5 or 10 U/mi), (B) monoclonal antibody IL-A 111 with or without IL-2, and (C) PMA (50 ng/ml) on host cell susceptibility. Lymphocytes were treated with the various reagents as described in Materials and Methods and then incubated with sporozoites for 30 min at 37 C before fixation. The infection index (shaded bars) and sporozoite internalization rate (hatched bars) are defined in the legend to Fig. 1. Results are given as the means ± standard deviations for three to five separate experiments. lymphocytes to infection by sporozoites after treating lymphocytes in vitro with agents that might be present at the tick attachment site. In all cases, the level of host cell susceptibility was increased between 30 and 125% compared with unstimulated controls. This increase in the number of sus- Control 100 U Interferon gamma FIG. 8. Effect of treating lymphocytes with gamma interferon (100 U/ml). The lymphocytes were incubated for 2 h at 37 C before the addition of sporozoites and further incubated for 30 min before the mixtures were fixed and processed for electron microscopy. The infection index (shaded bars) and sporozoite internalization rate (hatched bars) are defined in the legend to Fig. 1. Results are given as the means ± standard deviations for three separate experiments. ceptible cells within a population of lymphocytes occurred rapidly (less than 90 min when cells were stimulated with either tick salivary gland extract or IL-2). Furthermore, the increase in host cell susceptibility was not just an increase in the number of sporozoites bound to the cell surface; there was also a significant increase in the number of sporozoites fully internalized within individual susceptible cells. One interpretation of the latter observation is that there is an overall increase in the number of host cell surface molecules involved in the sporozoite entry process, although an increase in the avidity of the appropriate host cell surface receptor(s) or a clustering of these molecules cannot be excluded. The results of this study clearly demonstrate that treatment of lymphocytes with IL-2 prior to the addition of sporozoites produces a significant increase in host cell susceptibility. In contrast, the addition of IL-2 to T. parvainfected cells has a differential effect on parasite-host cell development depending on the time of application. For example, two recent studies have shown that IL-2 treatment of newly infected lymphocytes either had no effect on the establishment of infected cell cultures (5) or significantly suppressed the transformation and development of parasiteinfected cells (27). However, in both these reports, the addition of IL-2 to established macroschizont-infected cell lines resulted in enhanced proliferation of the cultures. Tick attachment site is an extremely favorable site for sporozoite infection. Although the exact site of sporozoite infection of lymphocytes in cattle has not yet been established, one obvious area is the lesion formed at the site of tick attachment. Because of the large population of macrophages in particular and the degranulation products of various leukocytes, along with possible adverse pharmacological and enzymatic effects due to tick saliva (28), it has been presumed that the tick-feeding lesion would represent a hostile environment for Theileria sporozoites (31). In contrast, we have provided some evidence that the tick attachment site might be a suitable site for infection. Shortterm (1 h at 37 C) incubation of lymphocytes with uninfected tick salivary gland extract, far from having a detrimental effect on either the lymphocytes or sporozoite invasion,

9 1494 SHAW ET AL. induces an increase in host cell susceptibility. Furthermore, since the tick attachment site is a region of inflammation, additional susceptible host cells may congregate within the region and cytokines, in particular IL-2, may be released locally, which may make those lymphocytes present even more susceptible to invasion by sporozoites. Whether a longer incubation with salivary gland extract would lead to a reduction in lymphocyte susceptibility was not investigated. However, a long (>6 h) exposure of lymphocytes to PMA results in a significant decrease in host cell susceptibility (unpublished results). The fact that the contents of the tick salivary glands might have a stimulatory effect on parasite invasion is, however, not unique. Previously, Titus and Ribeiro (30) described a similar phenomenon in the case of Leishmania transmission; salivary gland lysates administered with promastigotes significantly enhanced parasite infectivity in vitro. Similar results have also been reported for tick-borne arboviruses (15), suggesting that vector saliva may play an essential role in the establishment of a wide range of infections. Lastly, T. parva sporozoites can remain viable and infective in vitro for relatively long periods (up to 6 h when maintained at 37 C). If this is true for sporozoites in vivo, such longevity would appear to be counter to the possibility that the tick attachment site is the exclusive site of host cell invasion. There are at least two possible explanations for the relative longevity of isolated sporozoites. First, while the region around the tick attachment site may contain putative host cells, the number of susceptible cells present at any given time may be low. Therefore, a relatively long-lived sporozoite would increase the chances of the parasite coming into contact with and successfully invading potential host cells, as these cells may be continually moving into the lesion around the tick attachment site. Alternatively, a relatively long-lived sporozoite may eventually be swept via the draining lymphatic ducts into the efferent lymph nodes, where the chances of successful host cell invasion may be greatly increased by the large numbers of lymphocytes. ACKNOWLEDGMENTS We are grateful to Patrick Theuri, Daniel Ngugi, and Elias Awino for technical assistance, Maarten Sileghem for human recombinant IL-2 and bovine gamma interferon, Francis Mwakima and colleagues in the ILRAD Tick Unit for provision of ticks and sporozoites, and Molly Tilney for drawing the graphs. L. G. Tilney was supported by grant HD from the National Institutes of Health. REFERENCES 1. Alberts, B., D. Bray, J. Lewis, M. Raff, K. Roberts, and J. D. Watson Molecular biology of the cell, 2nd ed. Garland Publishing, Inc., New York. 2. Baldwin, C. L., S. J. Black, W. C. Brown, P. A. Conrad, B. M. Goddeeris, S. W. Kinuthia, P. A. Lalor, N. D. MacHugh, W. I. Morrison, S. P. Morzaria, J. Naessens, and J. Newson Bovine T cells, B cells, and null cells are transformed by the protozoan parasite Theileria parva. Infect. Immun. 56: Brown, C. G. D Theileriidae, p In A. E. R. Taylor and J. R. Baker (ed.), In vitro methods of parasite cultivation. Academic Press, London. 4. Cross, G. A. M Release and purification of clone-specific antigens constituting the surface coat of Trypanosoma brucei. Parasitology 71: DeMartini, J. C., and C. L. Baldwin Effects of gamma interferon, tumor necrosis factor alpha, and interleukin-2 on infection and proliferation of Theilena parva-infected bovine lymphoblasts and production of interferon by parasitized cells. Infect. Immun. 59: INFECT. IMMUN. 6. Duffus, W. P., G. C. Wagner, and J. M. Preston Initial studies on the properties of a bovine lymphoid cell culture line infected with Theileria parva. Clin. Exp. Immunol. 34: Emery, D. L., N. D. MacHugh, and J. A. Ellis The properties and functional activity of non-lymphoid cells from bovine afferent (peripheral) lymph. Immunology 62: Emery, D. L., N. D. MacHugh, and W. I. Morrison Theileria parva (Muguga) infects bovine T-lymphocytes in vivo and induces co-expression of BoT4 and BoT8. Parasite Immunol. 10: Fawcett, D. W., S. Doxsey, D. A. Stagg, and A. S. Young The entry of sporozoites of Theileria parva into bovine lymphocytes in vitro. Electron microscopic observations. Eur. J. Cell Biol. 27: Fawcett, D. W., and D. A. Stagg Passive endocytosis of sporozoites of Theileria parva in macrophages at 1-2'C. J. Submicrosc. Cytol. 18: Glass, E. J., E. A. Innes, R. L. Spooner, and C. G. D. Brown Infection of bovine monocyte/macrophage populations with Theileria annulata and Theileria parva. Vet. Immunol. Immunopathol. 22: Goddeeris, B. M., and W. I. Morrison Techniques for the generation, cloning, and characterization of bovine cytotoxic T cells specific for the protozoan Theileria parva. J. Tissue Culture Methods 11: Goddeeris, B. M., W. I. Morrison, A. J. Teale, A. Bensaid, and C. L. Baldwin Bovine cytotoxic T cell clones specific for cells infected with the protozoan parasite Theileria parva: specificity and class I major histocompatibility complex restriction. Proc. Natl. Acad. Sci. USA 83: Greenberg, S., J. El Khomy, F. Di Vingilio, E. M. Kaplan, and S. C. Silverstein Ca2"-independent F-actin assembly and disassembly during Fc receptor-mediated phagocytosis in mouse macrophages. J. Cell Biol. 113: Jones, L. D., E. Hodgson, and P. A. Nuttall Enhancement of virus transmission by tick salivary glands. J. Gen. Virol. 70: Lalor, P. A., W. I. Morrison, and S. J. Black Monoclonal antibodies to bovine leukocytes define heterogeneity of target cells for in vitro parasitosis by Theileria parva, p In W. I. Morrison (ed.), The ruminant immune system in health and disease. Cambridge University Press, Cambridge. 17. Lalor, P. A., W. I. Morrison, B. M. Goddeeris, R. J. Jack, and S. J. Black Monoclonal antibodies identify phenotypically and functionally distinct cell types in the bovine lymphoid system. Vet. Immunol. Immunopathol. 13: McKeever, D. J., N. D. MacHugh, B. M. Goddeeris, E. Awino, and W. I. Morrison Bovine afferent lymph veiled cells differ from blood monocytes in phenotype and accessory function. J. Immunol. 147: McKeever, D. J., and W. I. Morrison Theileria parva: the nature of the immune response and its significance for immunoprophylaxis. Rev. Sci. Tech. O.I.E. 9: Morrison, W. I., B. M. Goddeeris, W. C. Brown, C. L. Baldwin, and A. J. Teale Theileria parva in cattle: characterization of infected lymphocytes and the immune response they provoke. Vet. Immunol. Immunopathol. 20: Morrison, W. I., P. A. Lalor, B. M. Goddeeris, and A. J. Teale Theileriosis: antigens and host-parasite interactions, p In T. W. Pearson (ed.), Parasite antigens: toward new strategies for vaccines. Marcel Dekker, Inc., New York. 22. Moulton, J., G. Buscher, D. Bovell, and S. Doxsey Blast transformation of adherent macrophages infected in vitro with sporozoites of Theileria parva. Am. J. Vet. Res. 45: Naessens, J., J. Newson, A. Bensaid, A. J. Teale, J. G. Magondu, and S. J. Black De novo expression of T cell markers on Theileriaparva-transformed lymphoblasts in cattle. J. Immunol. 135: Naessens, J., M. Sileghem, N. MacHugh, Y. H. Park, W. C. Davis, and P. Toye Selection of BoCD25 monoclonal antibodies by screening mouse L cells transfected with the bovine ps5-interleukin-2 (IL-2) receptor gene. Immunology 76:

10 VOL. 61, 1993 INCREASED LYMPHOCYTE SUSCEPTIBILITY TO T. PARVA Norval, R. A. I., B. D. Perry, and A. S. Young The epidemiology of theileriosis in Africa. Academic Press, London. 26. Pinder, M., K. S. Withey, and G. E. Roelants Theileria parva parasites transform a subpopulation of T lymphocytes. J. Immunol. 127: Preston, P. M., C. G. D. Brown, and W. Richardson Cytokines inhibit the development of trophozoite-infected cells of Theilenia annulata and Theileria parva but enhance the proliferation of macroschizont-infected cell lines. Parasite Immunol. 14: Ribeiro, J. M. C Role of saliva in tick/host interactions. Exp. Appl. Acarol. 7: Shaw, M. K., L. G. Tilney, and A. J. Musoke The entry of Theileria parva sporozoites into bovine lymphocytes: evidence for MHC class I involvement. J. Cell Biol. 113: Titus, R. G., and J. M. C. Ribeiro Salivary gland lysates from the sand fly Lutzomyia longipalpis enhance Leishmania infectivity. Science 239: Walker, A. R Parasitic adaptions in the transmission of Theileria by ticks-a review. Trop. Anim. Health Prod. 22: Walker, A. R., and J. D. Fletcher Histological study of the attachment site of adult Rhipicephalus appendiculatus on rabbits and cattle. Int. J. Parasitol. 16: Webster, P., D. A. E. Dobbelaere, and D. W. Fawcett The entry of sporozoites of Theileria parva into bovine lymphocytes in vitro. Immunoelectron microscopic observations. Eur. J. Cell Biol. 36: