Abstract
Cryptococcus neoformans is a fungal pathogen that causes fatal meningitis and pneumonia. During host defense to Cryptococcus, NK cells directly recognize and kill C. neoformans using cytolytic degranulation analogous to killing of tumor cells. This fungal killing requires independent activation of Src family kinase (SFK) and Rac1-mediated pathways. Recognition of C. neoformans requires the natural cytotoxicity receptor, NKp30; however, it is not known whether NKp30 activates both signal transduction pathways or whether a second receptor is involved in activation of one of the pathways. We used primary human NK cells and a human NK cell line and found that NKp30 activates SFK → PI3K but not Rac1 cytotoxic signaling, which led to a search for the receptor leading to Rac1 activation. We found that NK cells require integrin-linked kinase (ILK) to activate Rac1 for effective fungal killing. This observation led to our identification of β1 integrin as an essential anticryptococcal receptor. These findings demonstrate that multiple receptors, including β1 integrins and NKp30 and their proximal signaling pathways, are required for recognition of Cryptococcus, which activates a central cytolytic antimicrobial pathway leading to fungal killing.
Introduction
Cryptococcus neoformans is a ubiquitous fungal pathogen that causes over 220,000 cases of meningitis per year and over 180,000 deaths annually (1). C. neoformans primarily infects AIDS patients. Current antifungal therapies have limited efficacy and are cost prohibitive in this population, and unfortunately, symptomatic individuals have a 10-wk mortality rate of 57% (2). This unacceptable morbidity and mortality has led to a search of mechanisms of host defense to Cryptococcus, with a view to enhanced therapies.
NK cells are innate immune cells that are best known for their role in killing of malignant and virus-infected cells. However, NK cells are also capable of killing microbes such as fungi (3). NK cell–mediated tumor killing depends on ligation of multiple receptors that trigger release of cytolytic proteins. When the signals from activating receptors and integrins dominate, the NK cell and the tumor target cell form a conjugate that creates an NK immune synapse (reviewed in Ref. 4). The synapse provides a platform that leads to intracellular signaling and trafficking of secretory lysosomes, resulting in directional release of cytolytic proteins toward the target cell. However, it is not known how nontumor ligands, such as those on fungal pathogens, trigger NK cell intracellular cytotoxic pathways.
Human and murine NK cells have been shown to directly kill C. neoformans (5–7). NK cells are recruited to the lung in response to pulmonary Cryptococcus (8), and mice lacking NK cells are more susceptible to cryptococcal infections (9). Once in contact, C. neoformans binds to the natural cytotoxicity receptor (NCR) NKp30 and initiates a cytotoxic PI3K → Erk1/2 signaling cascade (10). However, NKp30 does not directly interact with PI3K. Instead, cryptococcal stimulation activates both Rac1 and Src family kinase (SFK) upstream of PI3K. Activation of Rac1 and SFK are independent of one another, but both classes of proteins are necessary for the activation of the downstream cytotoxic PI3K → Erk1/2 pathway (11, 12). These observations triggered important unanswered questions. Does NKp30 activate two separate pathways (SFK and Rac1), and do these two separate pathways converge to activate PI3K, or does NKp30 activate one pathway and cooperate with other receptors to activate that other pathway?
During NK cell–mediated tumor killing, more than one receptor is involved. Both activating receptors and integrin signaling synergize to achieve cytotoxicity (13–15). Signaling through activation receptors leads to fusion of cytolytic granules with the plasma membrane (degranulation), whereas signaling through β2 integrins facilitates the movement of granules toward the immune synapse (granule polarization) (15). Both integrin-mediated granule polarization and activating receptor–mediated degranulation are essential to effective tumor lysis.
Although β2 integrins play a crucial role in tumor killing, they do not play a role in anticryptococcal activity (16). Although β2 integrins play the major role, β1 integrins have been implicated in tumor killing because cross-linking β1 integrins enhanced NK cell antitumor activity (17, 18). Interestingly, β1 integrins on neutrophils bind to and are activated by β-glucans, which are a conserved molecular pattern on fungal pathogens (19). Because Cryptococcus also expresses similar β-glucans, it is plausible that β1 integrins could be an anticryptococcal adhesion molecule.
To determine which signaling pathway NKp30 activated, we used blocking Abs to NKp30 and phospho-immunoblots to determine whether SFK or Rac is activated by this NCR. Having demonstrated that Rac is activated independently of NKp30, we used immunoblots following small interfering RNA (siRNA) knockdown or pharmacologic inhibitors to explore the role of integrin-linked kinase (ILK) upstream of Rac. Blocking Abs and siRNA knockdown were then used to investigate the roles of β2 and β1 integrins in activation of ILK → Rac and cryptococcal killing. Our results demonstrated that NK cells require two independent signaling pathways from NKp30 and β1 integrins. NKp30 activates SFK, and β1 integrins activate ILK. The SFK and ILK pathways then converge into a central PI3K signaling pathway.
Materials and Methods
Chemicals and small molecule inhibitors
FITC was purchased from Sigma-Aldrich (catalog no. 3326-32-7; St. Louis, MO), CPD 22 was purchased from EMD Millipore (catalog no. 407331; Etobicoke, ON, Canada). SMIFH2 was purchased from Calbiochem (catalog no. 344092). DMSO was obtained from Sigma-Aldrich (catalog no. 472301). Methyl-β-cyclodextran (MBCD) was obtained from Sigma-Aldrich (catalog no. C4555). CFSE was obtained from Millipore (catalog no. 4500-0270). Trypan blue stain 0.4% was purchased from Life Technologies (catalog no. 15250; Burlington, ON, Canada).
Abs
Proteins in immunoblots were revealed with specific Abs: rabbit anti–phospho-SFK (Y416) (2101S; Cell Signaling Technology, Whitby, ON, Canada), rabbit anti–phospho-Erk1/2 (T202/Y204) (9101S; Cell Signaling Technology), rabbit anti–p-ILK (AB1076; Millipore), mouse anti-Fyn (610163; BD Biosciences, San Jose, CA), mouse anti-Rac1 (1862341; Thermo Fisher Scientific, Waltham, MA), rabbit β1 integrin (4706S; Cell Signaling Technology), goat anti-rabbit IgG infrared dye 700DX (611-130-002; Rockland Immunochemicals, Limerick, PA), and goat anti-mouse IgG infrared dye 800 (923-32210; LI-COR Biosciences, Lincoln, NE). Cells for flow cytometry were labeled with specific Abs for mouse anti-CD11a PE-Cy5 (551131; BD Biosciences, San Jose, CA), rabbit anti-CD29 (4706; Cell Signaling Technology), and mouse anti-CD29 conjugated to Alexa Fluor 488 (303015; BioLegend, San Diego, CA).
Cells and Cryptococcus
YT cells are an NK-like cell line isolated from the pericardial fluid of a 15-y-old boy with acute lymphocytic leukemia (20). YT cells possessed NK cytotoxic activity against a panel of target cells, including K562, T, and B cell lines (20). Our YT cells were a gift from C. Clayberger, Emeritus Faculty, Stanford University, Stanford, CA. YT cells were validated by their expression of NKp30 and NKp44 and lack of CD3. YT cells were maintained in complete medium containing RPMI 1640 supplemented with 10% FBS (Invitrogen Life Technologies), 1% nonessential amino acids (catalog no. 11140; Invitrogen Life Technologies), 1% sodium pyruvate (catalog no. 11360; Invitrogen Life Technologies), and 1% penicillin–streptomycin in a 37°C 5% CO2 incubator.
K562 cells were obtained from American Type Culture Collection (ATCC) (catalog no. CCL243; Manassas, VA). 721.221 cells were purchased from ATCC (catalog no. CRL1855). Both K562 and 721.221 cells were maintained in complete media in a 37°C 5% CO2 incubator. Primary NK cells were isolated from healthy donors using an NK isolation kit (catalog no. 130-092-657; Miltenyi Biotec, San Diego, CA) as per the manufacturer’s instructions. Isolated cells were routinely >92% CD56+ and <0.5% CD3+. C. neoformans strain B3501 (catalog no. 34873; ATCC) was grown to log phase in Sabouraud dextrose broth (catalog no. 238230; Becton Dickinson, Mississauga, ON, Canada) on a 32°C shaker overnight.
Immunoblotting
YT cells (3 × 105–3 × 106) were preincubated with various inhibitors and their controls for 1 h in a 37°C CO2 incubator. YT cells were then coincubated with C. neoformans strain B3501 at an E:T ratio of 1:100 for various time points in a 37°C water bath. Cells were lysed in Nonidet P-40 lysis buffer containing 50 mM Tris (pH 7.4), 250 mM NaCl, 5 mM EDTA, 50 mM NaF, 1 mM Na3VO4, 1% Nonidet P-40, and 0.02% NaN3. Lysis buffer was supplemented with phosphatase (Roche, Mississauga, ON, Canada) and protease inhibitors (Roche). Lysates were separated on a 4–12% Bis-Tris NuPAGE gradient gel (catalog no. NP0335BOX; Invitrogen). After separation, samples were transferred to a nitrocellulose membrane and revealed with indicated Abs. Bands were recorded using an Odyssey infrared imaging system (LI-COR Biosciences). Densitometry was performed by measuring the area under the intensity plot using ImageJ (version 1.48; National Institutes of Health). The fold increase in signaling compared with unstimulated cells was calculated as (intensity of stimulated condition normalized to loading control) / (intensity of unstimulated condition normalized to loading control) − 1.
NK anticryptococcal killing assay
YT cells were cocultured with C. neoformans at an E:T ratio of 200:1 in round-bottom 96-well plates (catalog no. 163320; Thermo Fisher Scientific). CFU were determined at 24 and 48 h postinoculation. The anticryptococcal activity of primary NK cells was determined by coculture with C. neoformans at an E:T ratio of 100–400:1 in flat-bottom 96-well plates. Plates were centrifuged at 400 × g for 5 min to bring NK and Cryptococcus into contact. CFU were determined 24 h postinoculation. In experiments in which CPD 22 was used, the inhibitor was added to the YT or primary NK cells at the same time that Cryptococcus was added. Primary NK cell and YT cell viability was determined by trypan blue staining. Percent viability was calculated as follows: (number of trypan blue positive cells) / (total number of cells) × 100%. Concentrations of CPD 22 had minimal impact on viability of YT and primary NK cells.
Conjugate assay
C. neoformans strain B3501 was labeled following the procedure for C. gattii as previously described by Huston et al. (21). Briefly, C. neoformans was cultured overnight to the exponential phase of proliferation and labeled with 2.5 μg/ml FITC per 108 cells at 22°C for 10 min. C. neoformans was then washed three times with PBS. YT cells were labeled with 5 μl of anti-CD11a PE-Cy5 Ab for 30 min in a 37°C CO2 incubator. For the conjugate assay involving MBCD, 2 mM MBCD or PBS control was added immediately before incubation. YT cells and Cryptococcus were then incubated together for 10 min at 37°C in 200 μl of complete media. YT cells and Cryptococcus were agitated by pipetting. Conjugates were detected by Guava easyCyte Flow Cytometer (Cytosoft version 5.3; Guava Technologies, Millipore, Danvers, MA), and data were analyzed by FlowJo software (Tree Star, Ashland, OR). The percentage of NK cells in conjugates with C. neoformans was determined as follows: (number of green and red event) / (total number of red events) × 100%.
Rac-GTP precipitation
YT cells were unstimulated or stimulated with C. neoformans for 4 min in a 37°C water bath. YT cells were centrifuged at 6000 × g for 30 s, and the supernatant was decanted. YT cells were then lysed, and Cryptococcus was added to the unstimulated conditions to control for the additional volume of Cryptococcus in the stimulated conditions. Rac-GTP was then extracted according to manufacturer instructions with two modifications (17-10394; Millipore). A protease inhibitor (Roche) was substituted for leupeptin, and 5 μl of magnetic beads conjugated to the p21 binding domain (PBD) of p21-activated kinase (PAK) was used. Aliquots of whole-cell lysate were saved before the addition of PBD-coated beads so that the total levels of Rac1 could be determined. Active Rac, which had bound PBD-coated beads, was measured by immunoblot.
siRNA knockdown
siRNA specific against β1 integrin’s β-chain (CD29) (5′-GGAACCCUUGCACAAGUGA-3′) were purchased from Invitrogen Life Technologies and Thermo Fisher Scientific. siRNA against β2 integrin was purchased from Thermo Fisher Scientific. Nontargeting siRNA was purchased from Cell Signaling Technology. YT cells were transfected with 2 μg of siRNA specific against β1 integrin, β2 integrin, or control. A Nucleofector Kit V (Amaxa, Walkersville, MD) and Nucleofector II (Amaxa) were used to perform the transfection using the Nucleofector program O-017. YT cells were transferred into 500 μl of prewarmed complete media after transfection and then placed in a 37°C incubator for 10 min. Cells were then transferred to six-well plates and allowed to recover for 24 h in a 37°C CO2 incubator. The cells were then used in a killing assay as described above.
Statistics
GraphPad Prism (La Jolla, CA) was used to evaluate statistics. Error bars represent the SEM. Data were analyzed by one-way ANOVA with Bonferroni correction. A p value < 0.05 was considered to be a statistically significant difference. Percent reduction in conjugates compared with H2O or DMSO was analyzed by the column statistics program in GraphPad Prism. Different experiments were performed on different days using cells from different donors.
Ethics
Experimental protocols were approved and performed following the guidelines from the Conjoint Health Research Ethics Board of the University of Calgary (protocol number REB15-0600).
Results
Cryptococcus activates NKp30 and signals through SFK but not Rac1
Previous studies found that SFK and Rac1 were activated in NK cells independently and nonredundantly upstream of PI3K in response to Cryptococcus (12). Additionally, NKp30 was required for PI3K activation and cryptococcal killing (10). However, it was unknown whether SFK or Rac1 contributes to NKp30-dependent activation of PI3K. To determine whether SFK is activated downstream of NKp30, YT cells were cocultured with Cryptococcus in the presence of an inhibitory Ab to NKp30 (clone 1C01) (10), and SFK signaling was tested after cryptococcal stimulation (Fig. 1A, 1B). An immunoblot showed that YT cells stimulated with C. neoformans had increased phosphorylation of SFK. However, treatment with 1C01 caused a marked reduction in SFK activation in response to Cryptococcus compared with isotype control Ab, revealing that NKp30 recognizes Cryptococcus and activates the SFK pathway (Fig. 1A, 1B). Coincubation of NK cells with anti-CD56 did not inhibit SFK activation (Supplemental Fig. 1), suggesting that the SFK activation pathway is specific to NKp30. CD56 was chosen as a control because it is a canonical NK receptor that also leads to SFK activation (22, 23). To determine whether Rac is activated downstream of NKp30, the experiment was repeated, but Rac activation was assessed by precipitation with PBD of PAK. Rac was activated in response to C. neoformans; however, unlike SFK, Cryptococcus-induced Rac1 activation was not inhibited by the addition of inhibitory anti-NKp30 (Fig. 1C). This reveals that NKp30 does not stimulate the Rac1-mediated pathway and suggests that an unidentified NK receptor activates Rac1.
ILK is required for Rac1-dependent NK cell anticryptococcal response
NK cell tumor killing depends on activating receptors, such as NKp30, and integrins to optimize killing. Because integrins can activate Rac1 in NK cells, we investigated whether integrin signaling is involved in cryptococcal killing (17, 24). Because there are a large number of integrins on NK cells, we examined ILK as an indication of integrin signaling. We stimulated YT cells with Cryptococcus and investigated the activation of ILK by immunoblot. Immunoblot for phosphorylated ILK showed that stimulation with C. neoformans induced activation of ILK in YT cells (Fig. 2A, 2B). To determine the pathway downstream of ILK, the small molecule inhibitor CPD 22 was used to inhibit ILK activity. CPD 22 inhibits ILK activation and the phosphorylation of ILK targets. CPD 22 has been described as a specific inhibitor of ILK because the phosphorylation of the downstream targets was restored with the introduction of constitutively active ILK, and when tested at high concentrations against a panel of 20 kinases, CPD 22 did not significantly impact the activity of 19 of those kinases, was only 60% effective against P70S6K, and did not affect focal adhesion kinase (FAK) (25). CPD 22 was also specifically chosen because it was shown to inhibit the ILK→ PI3K pathway (25, 26). YT cells treated with CPD 22 and stimulated with Cryptococcus showed reduced Rac1 activation compared with control (Fig. 2C, 2D). This reveals that ILK activity is required for activation of the Rac1 pathway.
ILK activity is required for NK cell–mediated cryptococcal cytotoxicity
Because Rac1 is involved in cryptococcal killing (12) and ILK activates Rac1 (Fig. 2), we wished to examine if ILK was required for cryptococcal killing. We preincubated YT cells or primary NK cells with CPD 22 and measured anticryptococcal activity by assessing CFU. We found that YT cells and primary NK cells treated with CPD 22 showed reduced anticryptococcal activity (Fig. 3A, 3B). Taken together, these findings reveal that ILK and its signaling pathway play a crucial role in NK-mediated anticryptococcal killing.
β1 integrins but not β2 integrins are required for NK-mediated cryptococcal killing
NK cells express β1 and β2 integrins, although β2 integrins are the predominant integrins on NK cells that are involved in antitumor cytotoxic signaling (14, 27). β1 integrins are also known to activate ILK signaling (28). Because β2 integrins are not involved in anticryptococcal activity, we examined whether inhibition of β1 integrins inhibited anticryptococcal activity. To determine if β1 and not β2 integrins are involved in cytotoxicity, we reduced the expression of β1 and β2 integrins by siRNA (Fig. 4A, 4B). We found that YT cells with diminished expression of β1 integrins had reduced anticryptococcal activity, whereas YT cells with diminished β2 integrins experienced no change in anticryptococcal activity (Fig. 4C). The lack of β2 integrin involvement agrees with previous literature that showed that blocking Abs and siRNA knockdown of LFA-1 (CD11a/CD18) did not inhibit cryptococcal killing (16). To exclude potential off-target effects of siRNA knockdown, we repeated the cryptococcal killing assay with a different sequence of β1 integrin siRNA (Fig. 4D, 4E) that provided similar results. We also performed an anticryptococcal cytotoxicity assay in the presence of a blocking Ab against β1 integrins (anti-CD29). We found that blocking β1 integrins also reduced the anticryptococcal activity of YT cells and primary NK cells compared with control Ab (Fig. 4F, 4G). These findings show that β1 integrins are necessary for NK cell–mediated anticryptococcal killing, but β2 integrins are not.
β1 integrins are required to activate ILK in response to cryptococcal stimulation
Because β1 integrin and ILK are involved in anticryptococcal activity and β1 integrins are known to activate ILK (28), we examined if β1 integrins activated ILK in response to cryptococcal stimulation. Using siRNA knockdown of β1 integrin, we found that reduction in expression of CD29 caused a reduction in ILK activation in response to cryptococcal stimulation (Fig. 5A, 5B). This reveals that Cryptococcus activates β1 integrins to initiate an ILK → Rac1 signaling pathway.
β1 integrins not are required for conjugate formation
β1 integrins in NK cells can function as both adhesion receptors and enhancers of cytotoxic signaling pathways (17, 29). Our findings that inhibiting β1 integrins inhibits ILK activation could be due to β1 integrin activation of ILK and augmentation of the Rac1 cytotoxicity pathway, or β1 integrins could be enhancing adhesion between Cryptococcus and NK cells and assisting another receptor binding to Cryptococcus, which is then responsible for ILK activation. Adhesion mediated by β1 integrins is dependent on lipid rafts (30), and we found that disruption of lipid rafts with MBCD reduced NK cell–Cryptococcus conjugate formation (Fig. 6A). We then proceeded to investigate if knockdown of β1 integrins inhibited conjugate formation. Knockdown of β1 integrins reduced the fluorescent intensity from anti–β1 integrin Ab by 90%. We found that YT cells with reduced β1 integrin expression did not have reduced conjugate formation compared with nontargeting siRNA or mock transfected cells (Fig. 6B). This suggests that β1 integrins are not a major contributor to adherence between NK cells and Cryptococcus. Therefore, the inhibition of ILK signaling is likely the result of impaired β1 integrin signaling rather than reduced adherence. This is consistent with other studies that have shown that cross-linking β1 integrins on NK cells enhances their cytotoxicity (17).
Discussion
In this paper, we showed that 1) the anticryptococcal receptor NKp30 activates the SFK pathway but not the Rac1 signaling pathway, 2) β1 integrins are required for NK cell–mediated cryptococcal killing, 3) β1 integrins induce an ILK → Rac1 signaling pathway that is required for anticryptococcal activity, and 4) β1 integrins are not required for conjugate formation between NK cells and Cryptococcus. Together, these results describe a novel role of β1 integrins in activating NK cell–mediated anticryptococcal cytotoxicity.
We found that there are similarities but also important differences between fungal killing and tumor killing. Anticryptococcal killing has been shown to require both SFK and Rac1 signaling pathways (11, 12). We found that NKp30 is responsible for the activation of SFK (Fig. 1), and β1 integrins were found to activate the ILK→ Rac1 signaling pathway (Figs. 2, 5). This observation resembles tumor killing because both NK-activating receptors and integrin signaling are required. However, activation of Rac1 by β1 integrins represents a noncanonical pathway for integrin signaling.
Previous studies suggest a model of how signaling leads to Rac1 activation. β1 integrin–mediated activation of ILK is likely through the cytoplasmic portions of their β-chains (reviewed in Ref. 31). ILK then acts as a scaffold protein that contains ankyrin repeats, pleckstrein homology domains, and calponin homology domains (28). Parvins are a family of proteins that interact with ILK by binding to the calponin homology domain of ILK and are involved in smooth muscle contraction, actin microtubule attachment, cell polarization, and cell survival (32, 33). Interestingly, β-parvin has been shown to interact with the guanine nucleotide exchange factor α-PIX, which is an activator of Rac1 (34). Further work will be required to determine whether this pathway or some other is involved in NK cell killing of fungi.
Different classes of integrins signal differently in NK cell–mediated cryptococcal killing compared with tumor killing. In tumor killing, the β2 integrins, specifically LFA-1, serve as adhesion molecules and signaling proteins that lead to granule polarization (15). The impact of LFA-1 on granule polarization is independent of activation receptor signaling, such as from FcγR or NKp30 (15). In cryptococcal killing, NKp30 activates SFK (Fig. 1), and β1 integrins activate Rac1 signaling (Figs. 2, 5). SFK and Rac1 are both required to independently initiate a shared PI3K → Erk cytotoxic pathway. The convergence of receptor signaling demonstrates how activating receptors and integrin signals can synergistically enhance a single antimicrobial pathway. This is different from tumor killing, where activation receptors and integrins control independent components of cytotoxicity. LFA-1 also enhances NK cell adhesion to tumor targets (35), but during cryptococcal killing, β1 integrins are not required for adhesion because loss of β1 integrins did not affect conjugate formation (Fig. 6). These data demonstrate a novel role of β1 integrins in NK cytotoxicity.
The lack of involvement of β1 and β2 integrins in conjugate formation with C. neoformans (16) could explain why the conjugate between Cryptococcus and NK cells is weaker and its formation is delayed compared with tumor killing (36). This lack of an integrin-mediated adhesion could hinder cytotoxic signaling because it would be easier for the conjugate to be prematurely disrupted.
β1 integrins are not canonical direct cytotoxic receptors. Rather, they are typically involved in NK cell adhesion to the extracellular matrix (37). However, cross-linking of β1 integrins augments NK cell production of IFN-γ (38) and IL-8 (27). Cross-linking β1 integrins has also been shown to enhance Ab-dependent cellular cytotoxicity (17). In this paper, we have demonstrated that activation of β1 integrins by cryptococcal targets is necessary for the anticryptococcal activity initiated from NCR, which is similar to the function of LFA-1 in tumor killing. Because LFA-1 and β2 integrins are not involved in cryptococcal killing, it is tantalizing to consider that β1 and β2 integrins share redundancy in NK cell cytotoxicity. Future experiments can examine the potency of β1 integrin cytotoxic signaling compared with β2 integrins. Because NK antifungal activity is slower than tumor killing, β1 integrin cytotoxic signaling may be less potent than β2 integrins.
Our research has shown that cryptococcal stimulation activates β1 integrin signaling; however, the ligand responsible is unclear. β1 integrin ligands are mainly matrix proteins such as fibronectin and adhesion molecules such as VCAM1 (39). However, one study has shown that β1 integrins on polymorphonuclear leukocytes can bind to the fungal β-glucan PGG-glucan (40). Another possibility is that cryptococcal stimulation alters another NK cell surface molecule, which in turn stimulates β1 integrins in a cis interaction. In this case, knockdown of β1 integrins would not affect NK cryptococcal conjugate formations because β1 integrins are not binding to a cryptococcal ligand. β1 integrins have shown the capacity for forming cis interactions (41, 42), and NK cell cytotoxicity can be regulated by cis interactions (43). Therefore, future research will be needed to explore cryptococcal ligands or cis interactions that are required to activate β1 integrins.
In conclusion, β1 integrin activates a Rac-mediated cytotoxicity pathway required for NK cell–mediated killing of Cryptococcus. The finding that β1 integrin provides an activation signal rather than enhancing adherence highlights its novel role in fungal killing: it functions as a pathogen-associated molecular pattern receptor rather than an adhesion molecule.
Footnotes
This work was supported by a studentship from The Lung Association, Alberta & NWT (to R.F.X.). C.H.M. was supported by the Jessie Bowden Lloyd Professorship. This work was also supported by Canadian Institute for Health Research Grant 365812 (to C.H.M.).
The online version of this article contains supplemental material.
References
Disclosures
The authors have no financial conflicts of interest.