Research Article, J Immunol Tech Infect Dis Vol: 5 Issue: 3
Infection and Activation of Human Neutrophils with Fluorescent Leishmania infantum
| Davis RE1*, Thalhofer CJ2 and Wilson ME1,3,4* | |
| 1Interdisciplinary Program in Immunology, University of Iowa, Iowa City, IA, USA | |
| 2Agonox Inc., Portland, OR 97213 | |
| 3Departments of Internal Medicine and Microbiology, University of Iowa, Iowa City, IA, USA | |
| 4Veterans Affairs Medical Center, Iowa City, IA, USA | |
| Corresponding author : Mary E. Wilson
Department of Internal medicine and Microbiology, University of Iowa, Iowa City, IA, USA Tel: 319-356-3169 Fax: 319-353-4546 E-mail: mary-wilson@uiowa.edu |
|
| Received: April 01, 2016 Accepted: May 07, 2016 Published: May 10, 2016 | |
| Citation: Davis RE, Thalhofer CJ, Wilson ME (2016) Infection and Activation of Human Neutrophils with Fluorescent Leishmania infantum. J Immunol Tech Infect Dis 5:3. doi:10.4172/2329-9541.1000146 |
Abstract
Infection and Activation of Human Neutrophils with Fluorescent Leishmania infantum
Neutrophils (PMNs) are recruited in high numbers to sites of host infection by the protozoan parasites of the genus Leishmania. Although PMNs are capable of phagocytizing Leishmania parasites and are potent producers of antimicrobial compounds including reactive oxygen species (ROS), they are unable to control the establishment of infection. Prior studies document production of ROS in isolated PMNs incubated with Leishmania under conditions allowing phagocytosis, however without a measure of single cells’ responses it cannot be discerned whether PMN activation and ROS production is suppressed or ineffective in the cells that internalize the parasite. To address these interactions, we engineered a strain of fluorescent, mCherry-expressing Leishmania infantum (mCherry-Li).
Keywords: Leishmania ; Neutrophils; Flow cytometry
Keywords |
|
| Leishmania ; Neutrophils; Flow cytometry | |
Introduction |
|
| The Leishmania spp. is vector-borne protozoa that cause a spectrum of highly prevalent diseases of humans, referred collectively as leishmania sis. The most severe form of the disease is visceral leishmania sis caused, in the New World, by L. infantum in endemic regions [1]. Within its phlebotomine sand fly vector the Leishmania spp. assume an extracellular promastigote state, but soon upon inoculation into mammalian dermis the parasites are taken up by phagocytic cells wherein they transform to an obligate intracellular amastigote form. The mechanisms whereby the parasite is efficiently taken up by and survives within the mammalian phagocyte are critical to disease pathogenesis. Although parasites are found in macrophages through much of infection [2,3], it has become apparent that other cell types may perhaps act as the first haven for parasites within the host [4-8], but how these non-macrophage cell-types interact with Leishmania parasites is less-well understood. Many studies examine early Leishmania -host cell interactions by documenting in bulk culture [9]. However, methods examining the phenotypic responses of infected immune cells on a per-cell basis have been invaluable in increasing our understanding of early disease pathogenesis [5,10]. | |
| Neutrophils (PMNs), the most abundant circulating leukocyte in healthy humans, and are primary effector cells that provide host defense by migrating to sites of infection [11,12]. It is clear that PMNs play a role in Leishmania infections, but their effect on disease outcome remains unclear [5,6,8,13]. Neutrophils can internalize and kill at least some Leishmania promastigotes in vitro [14-18]. They are the first cell to migrate to, and take up parasites at the site of in vivo infection with either L. major [5,18-20] or with L. infantum [6] spp. Indeed, there appears to be distinct differences in the way various species of Leishmania result in activation and otherwise affect neutrophils [10,21-23]. In the context of disease progression, it has been suggested that, rather than control the infection, neutrophils exacerbate disease onset either by delaying apoptosis, by impairing activation of dendritic cells, or by serving as a vehicle route for silent entry of the parasite [4,24-26]. | |
| In order to investigate the interactions between L. infantum and infected mammalian cells, we engineered transgenic L. infantum expressing fluorescent proteins. We validated the use of mCherryexpressing L. infantum (mCherry-Li) to examine the infection of human PMNs in vitro by both fluorescent microscopy and flow cytometry. The detection of the mCherry signal also provided means to determine the temporal and dose-dependent uptake mCherry-Li by the PMNs. Furthermore, the ability to distinguish infected (mCherry+) from uninfected (mCherry-) PMNs allowed us to differentiate phenotypic responses based on infection status of individual. | |
Materials and Methods |
|
| Leishmania parasites | |
| A Brazilian strain of wild type L. infantum (MHOM/BR/00/1669) was maintained in hamsters by serial intra cardiac injection of amastigotes. Parasites were grown as promastigotes at 26°C in liquid hemoflagellate–modified minimal essential medium (HOMEM) [27]. The gene encoding either mCherry or dTomato [28] was cloned into Leishmania expression vector pIR1SAT (kindly provided by Stephen M. Beverly, Ph.D., Washington University at St. Louis). Wild-type parasite cultures were transfected with gel-purified integrating expression constructs by electroporation, and clones were selected from semisolid agar plates [29,30]. Transfectants, called mCherry-Li or dTomato-Li respectively, were grown to stationary phase in HOMEM. Genomic DNA was extracted from transfectants using a described procedure, subjected to restriction enzyme digestion with SmaI, and analyzed on a Southern blot [31]. Briefly, DNA digests were separated on a 0.8% agarose gel, transferred to Nylon (Roche, Mannheim,Germany) and UV cross–linked. The gene encoding streptothricin acetyltransferase (SAT) was amplified by PCR from pIR1SAT using the following primers: forward 5’-ATGAAGATTTCGGTGATCCC-3’, reverse 5’-TTAGGCGTCATCCTGTGCTC-3’. The probe was [32P]- labeled with a Random Primer Amplification Kit (Roche), incubated with the nylon filter in a roller bottle, and washed under high stringency conditions. | |
| Human neutrophils | |
| Protocols for all studies with human subjects were approved by the University of Iowa Institutional Review Board (protocol 201304709). Human neutrophils were isolated from venous blood of healthy adults after incubation in 3% Dextran (Pharmacosmos, Denmark) to sediment erythrocytes followed by density separation over a Histopaque Ficoll (Sigma-Aldrich, Missouri, USA) gradient, as previously described [32]. Purity and viability, assessed by Trypan blue (Invitrogen, California, USA) exclusion and microscopy, was consistently >95% viable neutrophils. | |
| Neutrophil infections for fluorescence microscopy | |
| For fluorescent microscopy, PMNs were labeled in 5 μm carboxy fluorescein succinimidyl ester (CFSE)(Molecular Probes-Invitrogen, California, USA) for 10 min, rinsed by centrifugation, and allowed to adhere for 30 minutes to 24-well plates containing coverslips that had previously been acid washed, flamed and coated with fibrinogen (10ug/ml of saline). Stationary phase mCherry-Li promastigotes were opsonized with 5% human serum for 15 minutes washed twice and cultured with PMNs at a multiplicity of infection (MOI) of 5:1. Infections were synchronized by centrifugation of plates (1400 rpm, 2 min), then incubated for varying times at 37°C prior to fixing in 2% paraformaldehyde. Images were generated using an Olympus IX-81 inverted fluorescence microscope. Images analyzed using Image J software. | |
| Neutrophil infection for flow cytometry and cytospin | |
| For flow cytometry, 2x107 unlabeled PMNs/ml were incubated with 1:1, 5:1 or 10:1 serum opsonized mCherry-Li in RPMI with 10% fetal bovine serum, 2 mM L-glutamine, and 100 μg streptomycin/ml. Control PMNs were incubated in buffer without parasites. Cells were incubated at 37°C, 5% CO2 for the time indicated in different experiments. To compare the detection of Leishmania parasite uptake as detected by flow cytometry vs. microscopy, Li parasites (either mCherry-expressing or CFSE stained) were incubated with PMNs in suspension as above for 30 minutes. PMN-parasites cultures were then divided into two portions, one analyzed by flow cytometry. The other portion was spun by cytospin onto a glass slide and cells stained with Hema 3 (modified Wright’s stain) and analyzed by light microscopy. Slides were prepared in duplicate. The numbers of internalized and surface-bound parasites were quantified with at least 200 cells counted per slide. | |
| Flow Cytometry | |
| For detection of reactive oxygen species and surface CD62L by flow cytometry, the same protocol for PMN infection was followed as above. After 15 minutes at 37oC, 5% CO2, DHR123 (Molecular Probes- Invitrogen, California, USA) was added to measure reactive oxygen species at a concentration of two micro molar (2uM). Five minutes later (20 minutes after adding parasites), neutrophils were put on ice and stained with antibodies. PMNs were suspended in FACS staining buffer (1x PBS, 0.5% HI FBS, 0.01% sodium azide) with 10% vol/vol normal rat serum to block Fc receptors. After washing, samples were incubated at 4°C with fluorochrome–conjugated primary antibody to human CD62L PE-Cy7 (BD Bioscience, California, USA) for 30-60 minutes. Cells were then fixed in 2% PFA prior to flow cytometric analysis. Samples were analyzed on a FACS LSR II (Becton Dickinson, California, USA) and data were analyzed using FlowJo software (Treestar Inc., Oregon, USA). | |
Results |
|
| Generation of recombinant mCherry-expressing Leishmania infantum | |
| In order to accurately detect L. infantum harboring cells, we generated a panel of transgenic L. infantum lines constitutively expressing different fluorescent proteins. Genes encoding mCherry or dTomato were introduced into the L. infantum by homologous recombination at the SSU ribosomal DNA locus using the vector pIR1SAT. Figure 1A illustrates the pIR1SAT-mCherry construct and insertion site. To determine whether pIR1SAT had been retained as an episome or had been integrated into the genome, Southern blots of L. infantum genomic DNA were probed with [32P]-DNA corresponding to the full-length streptothricin acetyltransferase (SAT) gene sequence. If pIR1SAT were not integrated, digestion of total promastigote DNA with the restriction enzyme SmaI should generate a 9.2 kb DNA fragment band due to linearizing the plasmid containing the SAT gene. In contrast, genomic integration of SAT and the fluorescence marker should generate a smaller fragment containing the SAT gene, digesting within the insertion and in the flanking SSU rDNA locus. Southern blot analysis revealed a 6.4 kb band in lanes containing genomic DNA from mCherry-Li and dTomato-Li transfectants not from wild type L. infantum , consistent with integration of transgenes into the parasite genome (Figure 1B). In contrast, a separate Southern blot of the pIR1SAT-mCherry plasmid digested with SmaI yielded the expected 9.2 kb band corresponding to the linearized plasmid (not shown). Fluorescent protein expression was verified by fluorescence microscopy. Isolated human PMNs were stained with CFSE (to simplify cytoplasmic visualization), allowed to adhere to a fibrinogencoated coverslip, and infected with mCherry-Li for 20 minutes. The majority of mCherry-Li associated with PMNs were engulfed by PMNs (Figure 1C). Internalized parasites could be differentiated from surface-adherent parasites microscopically (Figure 1D). These experiments suggest that mCherry-Li stably expressed the fluorescent marker and could, at least during short-term PMN-infections, continue to be detected even following phagocytic uptake of the parasites. | |
 |
Figure 1: Generating recombinant Leishmania expressing genes encoding fluorescent markers, using the pIR1SAT expression vector. (A) Schematic showing the mCherry pIR1SAT recombinant plasmid and the gene order upon into genome of L. infantum . dashed lines indicate the expected sites of homologous recombination. (B) Integration of fluorescent marker genes in parasites was confirmed by Southern blotting. Genomic DNA from transgenic dTomato, mCherry, and wild type L. infantum were digested with SmaI, separated on an agarose gel and incubated with [32P]â�?�?DNA probe to the streptothricin acetyltransferase (SAT) gene in pIR1SAT. (B,C) Confirmation that mCherry- expressing L. infantum express fluorescent protein after infection of neutrophils. PMNs isolated from a healthy donor were stained with cytoplasmic CFSE dye, allowed to adhere to fibrinogen-coated coverslips and then infected with mCherry-Li, pseudo colored red. Pictured are images of PMN with both internalized parasites (C) and the combination of internalized and adherent parasites (D) Scale bar is equal to 5 microns (�?µm). |
| L. infantum uptake by human neutrophils | |
| We examined the kinetics of attachment and uptake of L. infantum parasites by human neutrophils using flow cytometry. PMNs in suspension were incubated with mCherry-L. infantum at an MOI of 5:1, and we measured the uptake of the mCherry signal by flow cytometry (Figure 2A) with uninfected PMNs as the control. The proportion of mCherry+ PMNs increased over the first 80 minutes of incubation, indicating engulfment or surface-bound parasites associated with the PMNs. | |
|
Figure 2: mCherry-Li uptake by human neutrophils (A) Unopsonized mCherry-Li were incubated with human neutrophils at 37°C at an MOI of 5:1. After 20, 40 or 80 minute, cultures were harvested, fixed, and processed for flow cytometry. Uninfected PMNs were used to set the mCherry expression gate. (B) Increasing doses of human-serum opsonized mCherry-Li were incubated with human neutrophils for 30 minutes then fixed and analyzed by flow cytometry. Histograms represent increasing numbers of mCherry+ (infected) PMNs of total neutrophil population. MFI (mean fluorescence intensity) is indicated. (C) PMNs were exposed to mCherry-Li fixed, then analyzed, either by flow cytometry or by light microscopy to determine percentage of internalized parasites. Results are representative of two separate experiments. (D) Using CFSE-labeled Li parasites, the same protocol as above was performed, except that internalized and surface-bound parasites were quantified for each MOI. |
| PMNs were exposed for twenty minutes with varying MOIs of mCherry-Li. Higher MOIs resulted not only in higher percentages of mCherry+ PMNs, but an increased mCherry signal (MFI) indicating a higher parasite burden on a per-cell basis (Figure 2B). To determine whether mCherry+ PMNs detected by flow cytometry corresponded to PMNs with surface bound or internalized parasites, PMNs were cultured with mCherry-Li in increasing MOIs, fixed in 2% PFA, and divided into two portions. The first part was analyzed by flow cytometry (Figure 2C, black bars). The second part was applied to glass slides using a cytospin, stained and analyzed by light microscopy for the percentage of PMNs that clearly contained intracellular parasites (Figure 2C, white bars). We consistently detected similar but slightly higher numbers of infected PMNs by flow cytometry than microscopy at each MOI. To discern whether the difference was due to the numbers of surface bound, as opposed to internalized parasites, we analyzed the proportions of fluorescent CFSE-labeled parasites that were attached or internalized by microscopy, in parallel with flow cytometry (Figure 2D). At MOIs of 1:1 and 5:1 there was an insignificant trend, but at an MOI of 10:1 there was significantly more association of Li with PMNs measured by flow cytometry compared to microscopy. Again, there was a trend toward increased numbers of total (internalized plus surface-bound) PMN-associated parasites, but the difference did not reach statistical significance. This could indicate either those microscopic counts are underestimates, or that some parasites lysed and deposited their fluorescent dye in the recently infected neutrophils, as we previously reported [33]. With either scenario, these data suggest that the L. infantum : PMN ratio influences parasite burden in both individual cells as well as in the entire neutrophil population. Additionally, these results confirmed that the majority of PMNs marked with the mCherry signal contain internalized parasites, and that a minority have either attached parasites or recently killed intracellular Li parasites in our in vitro culture. | |
| PMN production of ROS and activation | |
| Studies examining interactions between PMNs and Leishmania species show that PMNs take up but do not eliminate all internalized parasites [4,25]. We questioned how internalization of L. infantum would alter the PMNs’ activation status and production of microbicidal compounds such as ROS. To address this question, we exposed human PMNs to mCherry-Li in suspension and used flow cytometry to distinguish between Li-associated or non-associated PMNs and to examine PMN ROS-production and activation. As a control for the independent effects of isolated procedures on PMNs, an unexposed group of PMNs that was derived from the same preparation was exposed to incubation conditions and buffer additions exactly in parallel with neutrophils infected with parasites or zymosan particles. Thus art factual effects of isolation procedures should be evident by inspecting this control. | |
| We measured oxidative burst with the fluorescent indicator 1,2,3- dihydrorhodamine (DHR123) [11]. After culturing mCherry-Li and PMNs at increasing MOIs, we compared DRH123 fluorescence in both infected (mCherry+) and “bystander (mCherry-) PMNs (Figure 3A). Parasite-containing (mCherry+) PMNs at each MOI showed increased ROS activity compared to uninfected bystanders from the same conditions. However, we noted that bystander neutrophils in both MOI 5:1 and MOI 10:1 both showed a substantial amount of DHR123 fluorescence. To test whether the oxidative response of bystander neutrophils was caused by a soluble mediator, we incubated neutrophils with supernatant from a previous neutrophil L. infantum infection (Figure 3B, bottom histogram) or from Leishmania - conditioned media (not shown). Neither supernatant induced ROS production from neutrophils. This suggests that stimulation of bystander neutrophils could require direct contact with parasites or with infected PMNs. | |
|
Figure 3: ROS production and activation by PMNs associated with mCherry-Li and â�?�?bystanderâ�? uninfected neutrophils. (A) Flow plots from a 30 minute infection of human neutrophils with opsonized mCherry-Li. (B) shows the expression of DHR123 in infected PMNs, indicating the amounts of ROS generated. PMNs were incubated with either mCherry-Li at the indicated MOI, or buffer, or PMN supernatant for 10 minutes at 37°C. DHR123 was then added to cultures, and after an additional 20 minutes at 37°C, cells were fixed and stained for flow cytometry. PMN supernatant was collected from a previous PMN infection (MOI 5:1, 30 minutes). (C) Flow cytometry plots of surface CD62L and DHR123 in mCherry+ or mCherry- PMNs incubated for 30 minutes with mCherry-Li at 37°C is shown. Negative and positive controls were PMNs incubated with buffer, or with opsonized zymosan (OZ), respectively. |
| The activation status of infected neutrophils was further explored by staining for surface CD62L (L-selectin), an integrin important for neutrophil migration through host endothelium, which is cleaved from the PMN cell surface upon activation [34,35]. Neutrophils were cultured with different ratios of mCherry-Li. The magnitude of activation, according to the loss of surface CD62L, and ROS generation was measured in mCherry-associated neutrophils within the culture (mCherry+ in top row of Figure 3C) or in bystander neutrophils (mCherry-; bottom row of Figure 3C). Both infected and bystander PMNs in the culture responded with ROS generation and CD62L loss, and the magnitude was impacted by the multiplicity of infection (Figure 3C). At MOIs of 5:1 and 10:1, the majority of mCherry+ PMNs produced DHR123 and was CD62L low. However, the bystander cells in these conditions showed large numbers of CD62Llow and DHR123+ cells, indicating large amounts of activation. Interestingly, the markers were not fully correlated: substantial numbers of CD62Llow cells remained DHR123 negative in all conditions, both in mCherry+ and mCherry- PMNs. Furthermore, at the 1:1 MOI, a group of mCherry+ neutrophils remained CD62Lhigh DHR123-. This requires additional validation, but could suggest that a subset of neutrophils is able to bind or internalize L. infantum without triggering either marker of neutrophil activation. | |
Discussion |
|
| It is well established that protozoan Leishmania spp. survive inside macrophages throughout chronic infection of their mammalian hosts [3]. Reports over the past 5 years have made it clear that neutrophils are among first cells to arrive at the site of a sand fly bite and to take up the parasite [5,6]. An essential step in our understanding of the role of PMNs in leishmania sis is the bidirectional effect of parasites on neutrophils and vice versa. Prior reports document generation of PMN ROS by the collective population of neutrophils in response to Leishmania phagocytosis in vitro [14,34]. Generation of transgenic parasites expressing fluorescent markers enabled us to examine these questions on a single cell basis. We were thus able to study the interactions between human PMNs from healthy human donors and L. infantum promastigotes. | |
| Our investigations confirmed that human PMNs take up serumopsonized mCherry-Li promastigotes in vitro , as detected by both microscopy and flow cytometry. Parasite burden in neutrophils was increased by longer incubation times, peaking at 80 minutes, and also by incubation with greater ratios (MOIs) of parasites. Using the mCherry marker to distinguish between infected and uninfected neutrophils in the culture, we found that infected neutrophils showed greater levels of ROS and enhanced activation (evidence by loss of surface CD62L) compared to unexposed neutrophils. However, this response was heterogeneous, and this was particularly evident at low infection rates. Subsets of infected (mCherry+) neutrophils avoided acquisition of both markers of activation (ROS production and loss of CD62L). This finding contrasts with early reports homogeneous activation of PMNs in response to parasites in vitro [14]. | |
| Previous reports were not able to differentiate whether ROS production corresponded to cells harboring intracellular parasites, and thus allows us to distinguish the responses of individual infected or bystander cells. Our observation that neutrophil populations exhibit heterogeneous responses to parasites in culture is consistent with observations in vitro and in vivo that PMNs take up, and though they may kill some parasites early in infection [13], they are unable to control Leishmania infection [5,22,23,25]. Our data raise the curious possibility of “silent” parasite uptake by a neutrophil subset. It is possible that there could be a threshold of infection level required to stimulated ROS production by PMNs [14,34,36], and/or that parasites may prevent ROS generation by some PMNs, similar to their inhibitory effects on some macrophage functions [37]. It has long been established that Leishmania parasites can bind and utilize surface receptors on macrophages (including CR3, CR1 and mannose receptor) as a means of “safe entry” [38-40]. Of these receptors, only CR3 (a.k.a. CD11b/CD18) is highly expressed on PMNs [38]. CR3 on macrophages does not trigger activation, and it is possible that neutrophil CR3 is also involved in suppression of neutrophil responses in a subset of cells. | |
| Despite observations of murine neutrophils to the contrary [21,23], our data also showed activation of “bystander” human PMNs in cultures containing L. infantum infected neutrophils. Bystander activation was not observed when fresh PMNs were incubated in conditioned medium from either L. infantum or L. infantum infected neutrophils. This leads us to speculate that transient physical interactions with parasites, or with parasite-infected or activated neutrophils, may trigger activation of uninfected bystander neutrophils. Finally, in both infected and uninfected PMNs the loss of CD62L and production of ROS do not always occur in concert, suggesting differences between the triggers of both these activation associated events. | |
| The contribution of neutrophils to the control or exacerbation of chronic leishmania sis remains controversial, in part due to different systems of study and lack of specificity of depletion methods [13,41,42]. One thing that is clear is that neutrophils are abundant and interact extensively with the parasite during the first hours of infection [5,6,19]. Many other diseases such as those causing lung inflammation [43] or autoimmune disorders [44] document a role for highly activated neutrophils contributing to off-target cellular pathology. Our findings suggest that the concentration of parasites during the infection of infiltrating PMNs influences not only infected, but also bystander PMNs. We hypothesize that the infected PMNs responding to L. infantum phagocytosis with low or absent ROS production and/or activation could promote the maintenance of a parasite subset that establishes infection. Second, responses of some infected and of bystander PMNs may contribute to a local inflammatory environment that is ineffective at parasite clearance. There may be a larger role for PMNs in establishing infection and modulating the local immune environment at the inoculation site than has previously been appreciated. | |
Acknowledgments |
|
| This was supported in part by grants AI045540 and AI076233 from the NIH and Merit Review grants 5I01BX001983 and 2I01BX000536 from the Department of Veterans’ Affairs. The work was performed while RD received support from NIH Immunology Training Grant 5T32AI07485. | |
References |
|
|
|
