Early Production of the Neutrophil-Derived Lipid Mediators LTB4 and LXA4 Is Modulated by Intracellular Infection with Leishmania major

Recruitment of neutrophil granulocytes to sites of infectious tissue damage is an early event in innate immune responses. Following chemotactic signals neutrophils establish a first line of defense in a swarm-like manner. Intracellular pathogens such as Leishmania major can, however, evade neutrophil-mediated killing and survive inside neutrophils. To achieve this the parasites evolved potent evasion mechanisms. Since neutrophils are a major source of inflammation regulating lipid mediators, we hypothesized that intracellular infection modifies the release of pro- and anti-inflammatory lipid mediators like leukotriene B4 (LTB4) and lipoxin A4 (LXA4), respectively. In the present study, we demonstrated in vitro that L. major-infected primary human neutrophils release an increased amount of LTB4, whereas LXA4 liberation is reduced during the first hours of infection. To investigate whether lipid mediator modulation is a common feature in intracellular infections, we tested the impact of an infection with Anaplasma phagocytophilum. Similarly to L. major, neutrophil infection with A. phagocytophilum led to an enhanced release of LTB4 and decreased LXA4 production. Together, our findings indicate that intracellular infections modulate the lipid mediator profile of neutrophils. This effect is likely to contribute to the survival of the pathogens in neutrophils and to the outcome of the infections.


Introduction
As the most numerous cell type in early acute inflammation, polymorphonuclear neutrophil granulocytes (PMN) can be central choreographers of inflammation [1][2][3][4]. During this phase, they occupy an outstanding position in the regulation of local lipid mediators, a family of mainly arachidonic-acidderived signal molecules with potent effects on leukocyte recruitment and activity [5,6]. The proinflammatory lipid mediator leukotriene B4 (LTB 4 ) works as an amplifier for localized inflammatory signals and was shown to be critical for sufficient PMN recruitment, termed swarming, in vivo [7]. In addition, LTB 4 was shown to enhance phagocytic activity, activation, degranulation, and killing of internalized pathogens [8][9][10][11]. Since PMN are the predominant source of LTB 4 , they promote acute inflammation in a feedforward manner [12,13]. In addition to their proinflammatory functions neutrophils also contribute to the resolution of inflammation. To inhibit overwhelming inflammation, the LTB 4 -precursor LTA 4 is used for the synthesis of proresolving lipid mediators (SPMs). This process, termed class switch, is characterized by the production of lipoxin A 4 (LXA 4 ), a prototype member of SPMs [5,14]. LXA 4 is known to inhibit leukocyte chemotaxis, transmigration, ROS-generation, NF-kB activation, and synthesis of proinflammatory cytokines [15,16]. LXA 4 is mainly produced by transcellular dual lipoxygenation. In this process PMN-derived LTA 4 is used as a substrate of 12-and 15-LO expressed in epithelial cells [17,18]. Therefore, neutrophils not only produce the proinflammatory lipid mediator LTB 4 but also contribute to the production of the anti-inflammatory mediator LXA 4 . Upon exposure to the calcium ionophore ionomycin the LTB 4 and LXA 4 production by neutrophils can be regarded as a measurement of the cells' capacity for the synthesis of these lipid mediators. By using LPS + fMLP information regarding the LTB 4 release by neutrophils in an infected/inflammatory environment can be obtained.

BioMed Research International
Certain pathogenic microorganisms such as the protozoan parasite Leishmania major (L. major) can evade destruction and survive inside neutrophils [19]. L. major is an obligatory intracellular parasite and causative agent of Old World cutaneous Leishmaniasis. During blood meal the parasites are transmitted into the skin of mammalian hosts by the bite of infected sandflies. Neutrophils are rapidly recruited to the site of Leishmania infection and phagocytose the parasites [20,21]. In this context, L. major parasites were shown to exploit the early inflammatory response by using PMN as transient host cells [22][23][24][25]. Lipid mediators have been shown to be involved in the survival strategy of Leishmania parasites [26][27][28][29]. However, no data concerning the impact of L. major infection on neutrophil-derived lipid mediators is available.
In the present study, we investigated how intracellular infection with L. major affects the release of the pro-and anti-inflammatory lipid mediators LTB 4 and LXA 4 by primary human neutrophil granulocytes in vitro. In addition to L. major, we tested the impact of an infection with Anaplasma phagocytophilum (A. phagocytophilum) on LTB 4and LXA 4 -synthesis to investigate whether the neutrophil lipidome is a common target of intracellular pathogens. A. phagocytophilum is an obligate intracellular bacterium and causative agent of the tick-borne Human Granulocytic Anaplasmosis (HGA). It is critically dependent on neutrophils as its definitive host cells and well known for host cell modulations that lead to subversion of PMN antimicrobial defense mechanisms to ensure intracellular survival [30].

Ethics.
Blood collection was conducted with the understanding and written consent of each participant and was approved by the ethical committee of the Medical Faculty of the University of Lübeck (05-124).

Leishmania Major
Culture. The origin and propagation of the cloned virulent L. major strain MHOM/IL/81/FEBNI has been described elsewhere [32]. In short, L. major promastigotes were cultured on biphasic rabbit blood agarcontaining microtiter plates at 26 ∘ C in humidified atmosphere containing 5% CO 2 for 7 to 10 days and a maximum of five passages. Each plate well contained 100 l liquid medium and 50 l of a Novy-MacNeal-Nicolle (NNN) blood agar slant, which was prepared by supplementing 200 ml of Brain-Heart-Infusion agar base (Difco, Detroit, MI, USA) with 50 ml defibrinated fresh rabbit blood (Elocin-Lab, Oberhausen, Germany). The liquid medium consisted of RPMI-1640 medium supplemented with 5% heat-inactivated fetal calf serum and 25 M -mercaptoethanol (all from Sigma-Aldrich), 4 mM L-glutamine, 10 mM HEPES, 100 U/ml penicillin, and 100 g/ml streptomycin (all from Biochrom). For infection L. major promastigotes were washed with complete medium for 10 minutes at 2600 ×g. After centrifugation the supernatant was discarded and the pellet was resuspended in complete medium. Promastigotes with active flagellar movement were counted in a hemocytometer with a chamber depth of 0.02 mm.

Preparation of Cell-Free Anaplasma phagocytophilum.
The A. phagocytophilum Webster strain was a kind gift of Dr. J. S. Dumler, John Hopkins University, Baltimore, MD. The bacteria were propagated and cell-free A. phagocytophilum was prepared as described previously [33]. Briefly, infected HL-60 cells were centrifuged at 250 ×g for 10 min and resuspended in 2 ml of PBS. Subsequently, cells were passed through a 25 G needle followed by a 27 G needle for 10 times each and vortexed with sterile solid glass-beads for 1 minute. By centrifuging at 750 ×g for 10 min cellular debris was removed. The supernatant was collected and centrifuged at 2.500 ×g for 15 min. The obtained cell-free Anaplasma containing pellets were used to infect neutrophils.

HT-29 Cell
Culture. The human adenocarcinoma epithelial-like cell line HT-29 was kept in DMEM medium supplemented with 10% heat-inactivated FCS (both Sigma-Aldrich), 2 mML-glutamine, 100 U/ml penicillin, and 100 g/ml streptomycin (all from Biochrom) under humidified conditions at 37 ∘ C and 5% CO 2 . Placed in a well of a 96-well flat bottom microplate 9 × 10 4 cells formed a confluent monolayer within 24 hours. Preliminary experiments showed that HT-29 cells do not release LTB 4 nor LXA 4 on their own (data not shown).

In Vitro Coincubation of Neutrophils with Pathogens and
Determination of Infection Rate. Neutrophils (5 × 10 6 per ml) were coincubated with L. major promastigotes or cell-free A. phagocytophilum in complete medium for 300 min at 37 ∘ C in humidified atmosphere containing 5% CO 2 . The multiplicity of infection (MOI) for L. major was 5. For infection with A. phagocytophilum the infectious load was 1 : 1 meaning that one neutrophil was infected with A. phagocytophilum obtained from one infected HL-60 cell. Subsequently, neutrophils were washed three times (400 ×g, 10 min) with PBS to remove noningested pathogens. Immediately before induction of lipid mediator release, the cells were resuspended in FCS-free complete medium since preliminary studies indicated a high FCS induced background signal in the LTB 4 -ELISA assay. The infection rates for L. major and A. phagocytophilum were determined by morphological examination of >200 PMN after Diff Quik staining of cytocentrifuge slides. A. phagocytophilum bacteria were visualized in neutrophils by immunocytochemical staining with the use of a polyclonal anti-A. phagocytophilum antibody (a kind gift of Professor J. Stephen Dumler, John Hopkins University, Baltimore, MD). 4 . Prior to the induction of LTB 4 release, the cells were resuspended in FCS-free complete medium. Infected and noninfected PMN as well as pathogen controls (L. major promastigotes or cell-free A. phagocytophilum without neutrophils) were exposed to either the combination of LPS and fMLP (1 g/ml LPS for 30 min followed by 0.5 M fMLP 10 min; both Sigma-Aldrich) or ionomycin (0.2 M, 10 min, Sigma-Aldrich). Control samples were left untreated. The induction was carried out in a 37 ∘ C water bath and terminated by cold centrifugation (4 ∘ C, precooled, 800 ×g, 10 min). The supernatants were collected and stored at −80 ∘ C. LTB 4 was finally quantified by competitive enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instructions. 4 . Neutrophils were coincubated with L. major or with A. phagocytophilum for 300 min as described above. For the induction of LXA 4 10 ng/ml GM-CSF (Peprotech, Hamburg, Germany) was added as priming agent [34] for the last 90 minutes of incubation. The same procedure was applied to the pathogen controls. All cells were then washed and resuspended in FCS-free complete medium. The cells were then transferred to a 96-well flat bottom microplate with wells containing a confluent monolayer of HT-29 epithelial cells. The ratio of HT-29 cells to PMN in the coculture was approximately 1 to 7.5. For the induction of LXA 4 , infected and noninfected PMN as well as pathogen controls were treated with 1 M ionomycin (10 min or 360 min; Sigma-Aldrich) or left untreated. Stimulation was carried out at 37 ∘ C in humidified atmosphere containing 5% CO 2 and terminated by centrifugation at 4 ∘ C (precooled, 800 ×g, 10 min). Supernatants were collected and stored at −80 ∘ C. LXA 4 was finally quantified by ELISA (United States Biological, Salem, MA, USA) according to the manufacturer's instructions.

Statistical
Analysis. Student's t-test for paired samples was performed in GraphPad Prism 7 software (La Jolla, CA, USA) for the comparison of lipid mediator content in the supernatants of infected and noninfected PMN after induction with ionomycin or LPS + fMLP. Data are presented as mean (±SD). Differences with a value ≤ 0.05 were considered significant.

Infection with Leishmania major Leads to Enhanced
Production of LTB 4 by Primary Human Neutrophils. Primary human neutrophils were infected in vitro with stationary phase L. major promastigotes. After five hours of coincubation with the parasites the ratio of infected neutrophils was 65% ± 15% as determined by examination of Diff Quikstained slides (Figure 1(a)). Treatment with ionomycin or the combination of LPS and fMLP resulted in the release of LTB 4 (Figures 1(b) and 1(c)). Infection with L. major significantly ( = 0.048) enhanced the ionomycin-induced LTB 4 release from 11.5 (±3.4) ng/ml to 16.0 (±2.1) ng/ml (Figure 1(b)). After infection with L. major the LTB 4 -release induced by LPS + fMLP was also significantly ( = 0.049) enhanced from 0.24 (±0.23) ng/ml to 0.48 (±0.37) ng/ml (Figure 1(c)). The infection with L. major itself, without additional stimulation, did not result in the release of LTB 4 (Figures 1(b) and 1(c)).

Discussion
Lipid mediators are essential regulators of neutrophil granulocyte recruitment and activity [13,35]. In this study we investigated the influence of intracellular infections on the early release of LTB 4 and LXA 4 by primary human neutrophils in vitro. We showed that infection with either L. major or A. phagocytophilum leads to an increased release With regard to the neutrophil-dependent establishment of productive L. major infections and the Trojan horse hypothesis [22,23] our findings suggest that L. major parasites abuse infected PMN for augmented recruitment of neutrophils. By increasing neutrophil-derived LTB 4 as a key recruitment factor [7] along with an upregulated IL-8 [36], L. major can ensure the sufficient presence of transient host cells for the subsequent infection of macrophages [22]. In accordance with this view, we could show a downregulation of the recruitment antagonist LXA 4 during the first six hours of infection. As L. major parasites benefit from high doses of LXA 4 in the later phase of inflammation [37], it seems likely that in the beginning of an infection an increased PMN recruitment induced by LTB 4 has a higher priority in the Leishmania survival strategy than the downregulation of PMN effector functions by LXA 4 .
Previously, our group identified a Leishmania promastigote-derived lipid mediator termed Leishmania chemotactic factor (LCF) [36]. On the one hand, comparable to LTB 4 , this lipid selectively recruits PMN and induces PMN-derived IL-8 which forms an amplifying loop for neutrophil recruitment [36,38]. On the other hand, LCF  to Leishmania major, the observed increase of PMN-derived LTB 4 and decrease of LXA 4 production could enable the recruitment of large numbers of host cells and are likely to contribute to the establishment of A. phagocytophilum infection.
Since early after infection the survival of both pathogens depends on the recruitment of sufficient number of host neutrophils, the enhanced induction of LTB 4 production appears to be crucial to successfully establishing the infection of the host. Moreover, the enhancing effect of LTB 4 on the neutrophil phagocytic capacity [39] likely contributes to the entry of the pathogens into neutrophils. However, LTB 4 also leads to enhanced antimicrobial effector functions such as the production of reactive effect oxygen species and phagosomelysosome fusion [40,41]. Since these functions can be detrimental for both L. major and A. phagocytophilum, the pathogens must possess effective mechanisms to evade these antimicrobial effector functions. Indeed, both L. major and A. phagocytophilum can effectively inhibit both the production of ROS [42,43] and acidification of the phagolysosomes [30,44].
For their transmission both Leishmania and Anaplasma are dependent on vectors. Salivary gland extracts (SGE) of Lutzomyia longipalpis sandflies that are insect vectors for Leishmania parasites were found to inhibit LTB 4 production and LTB 4 -mediated chemotaxis [45]. Furthermore, SGE induces PGE 2 which in turn can promote LXA 4 synthesis. Consequently, SGE was shown to facilitate survival of L. infantum [46]. Similar to sandflies, Ixodes ricinus, the tick vector of A. phagocytophilum, secretes a leukotriene binding protein termed Ir-LBP that works as a "scavenger" for LTB 4 . By this effect tick saliva can decrease the number and activation level of PMN located at the tick bite site in vivo [47]. In summary there seems to be opposite LTB 4signaling of vector and pathogen due to different intentions. Whereas sandflies and ticks aim to secure blood meals by dampening inflammation, the pathogens need inflammation for sufficient host cell influx.
Other intracellular pathogens that are not dependent on neutrophils as host cells show different approaches to influencing the balance of lipid mediators. For example, Mycobacterium tuberculosis inhibits proinflammatory PGE 2 and enhances lipoxin synthesis in vivo [48]. Toxoplasma gondii has been shown to express its own 15-LOX leading to an increased lipoxin synthesis [49]. By induction of antiinflammatory lipid mediators these pathogens seem to differ from the group of intracellular pathogens that use neutrophils as host cells. Nevertheless, these findings in concert with our data illustrate that manipulation of local lipid mediators by pathogens is a widespread phenomenon.

Conclusions
Taken together, our data support the view [50] that the neutrophil lipidome is a common target of intracellular parasites to modulate the influx and function of leukocytes at the site of infection. We showed that the vector-transmitted intracellular pathogens Leishmania major and Anaplasma phagocytophilum promote neutrophil-driven amplification of acute inflammation. This modulatory function likely contributes to the recruitment of host cells that are essential for the survival and multiplication of the pathogens. Further studies and in vivo models should confirm the impact of intracellular pathogens on host neutrophil lipidome. This could in turn provide a basis for therapeutic approaches to counteract the pathogens survival strategies at this level.

Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this article.