HIF2α is a Direct Regulator of Neutrophil Motility

Orchestrated recruitment of neutrophils to inflamed tissue is essential during initiation of inflammation. Inflamed areas are usually hypoxic, and adaptation to reduced oxygen pressure is typically mediated by hypoxia pathway proteins. However, it is still unclear how these factors influence the migration of neutrophils to and at the site of inflammation either during their transmigration through the blood-endothelial cell barrier, or their motility in the interstitial space. Here, we reveal that activation of the Hypoxia Inducible Factor-2 (HIF2α) due to deficiency of HIF-prolyl hydroxylase domain protein-2 (PHD2) boosts neutrophil migration specifically through highly confined microenvironments. In vivo, the increased migratory capacity of PHD2-deficient neutrophils resulted in massive tissue accumulation in models of acute local inflammation. Using systematic RNAseq analyses and mechanistic approaches, we identified RhoA, a cytoskeleton organizer, as the central downstream factor that mediates HIF2α-dependent neutrophil motility. Thus, we propose that the here identified novel PHD2-HIF2α-RhoA axis is vital to the initial stages of inflammation as it promotes neutrophil movement through highly confined tissue landscapes.


Introduction 50
In the innate immune response, neutrophils represent the first line of protection against 51 infections, extravasating quickly from circulation to inflamed tissues for fast pathogen 52 elimination. This process necessitates transit from an oxygen-rich circulatory system to the 53 inflammation site, which is typically hypoxic due to vasculature damage and/or high metabolic 54 demand of pathogens and host cells. 1 Thus, neutrophil adaptation to low oxygen levels is 55 crucial during the early phases of the inflammatory response. 56 Under hypoxic conditions, the transcription factors Hypoxia Inducible Factor-1 (HIF1α) and 57 its isoform HIF2α are key elements that control immune cell metabolism and function, 2-7 and 58 importantly, HIF activity is controlled by a class of oxygen sensors known as the HIF prolyl-59 hydroxylase domain enzymes (PHD1-3) (reviewed in 8,9 ). When oxygen levels decrease, PHDs 60 get inactivated, which results in HIFα stabilization and transcription of relevant target genes. 61 Interestingly, HIF1α-deficiency results in subdued inflammation 2,3 while, inversely, PHD 62 inactivation and/or HIFα stabilization leads to enhanced neutrophil survival, 4,10 chemotaxis and 63 degranulation (reviewed in 11 ). Although both HIFα subunits have overlapping activities, 64 unique roles for HIF2α, including in neutrophil function, have been reported. 4-6 65 Over the past decade, several mechanisms have been shown to participate in the multi-step 66 recruitment of neutrophils from circulation to sites of infection or inflammation. [12][13][14] The 67 recruitment process requires cell plasticity because cells deform as they move through the 68 blood-endothelial cell barrier and the confined areas of interstitial tissues. Leucocyte migration 69 through these microenvironments is orchestrated by actin polymerization regulators, such as 70 the rho GTPases RhoA, Cdc42 and Rac1. [15][16][17][18] In this context, HIF1α expression has been 71 suggested to modulate both functional changes in the cytoskeleton and metabolic reprograming 72 [19][20][21][22] . Importantly, disruption of mechanisms that control neutrophil infiltration in tissues is 73 7 After polymerization of the collagen matrix, phase imaging was performed using video-144 microscopy (DMI8). For the chemotaxis assay, CXCL2 (20ng/ml) Figure 1A). Initially, 1D migration assays in polydimethylsiloxane (PDMS) 173 micro-channel devices of different levels of constriction (channel widths of 3, 4 or 5µm) were 174 used to characterize the migratory capacity of individual neutrophils ( Figure 1A). 17,18,31-34 175 Interestingly, cKO P2 neutrophils moved significantly faster than their WT counterparts, but 176 only in the most confined channels ( Figure 1B and Supplemental Figure 1B). To identify 177 downstream effectors of this phenotype, we evaluated the contributions of HIF2α, a PHD2 178 target and a central factor in inflammation 4,35 , in cKO P2H2 neutrophils compared to their 179 littermate controls (Supplemental Figure 1C). Interestingly, there were no differences in speed 180 at any of the degrees of confinement tested ( Figure 1C). These data strongly suggest that 181 enhanced HIF2α activation regulates neutrophil motion in very confined microenvironments. 182

183
We extended our analysis to evaluate neutrophil migration in a 2D confined microenvironment 184 (4.5 µm height) ( Figure 1D). Similar to the results obtained in the 1D migration assay, 185 neutrophils from cKO P2 mice showed increased motility compared to their WT counterparts, 186 as evidenced by longer trajectories of equivalent durations ( Figure 1E), as well as greater speed 187 ( Figure 1F), along with higher mean square displacement (MSD) values ( Figure 1G). On the 188 other hand, under identical conditions, cKO P2H2 neutrophils did not show any difference in 189 speed or MSD compared to their WT counterparts ( Figure 1H and 1I). Interestingly, cell 190 migration in a non-confining 2D chamber (12µm height) showed no difference in speed, 191 trajectories, or MSD (Supplemental Figure 1D-F). Thus, these data indicate that the  HIF2α pathway regulates cell migration by facilitating mobility strictly in confined spaces. 193 194 PHD2-deficient neutrophils display enhanced non-directed motility in complex 195 environments 196 We used 3D-collagen matrices to confirm the role of PHD2 in neutrophil migration in a 197 microenvironment of fibers and different pore sizes, adequately mimicking the tissue 198 complexity in vivo. Therefore, migration of freshly-isolated BMDNs from cKO P2 mice and 199 WT littermates was compared in dense 3D collagen gels (4mg/ml) ( Figure 2A As it has been suggested that silencing of PHD2 in neutrophils leads to their enhanced 207 chemotaxis, 36 we assessed this effect in our complex 3D collagen matrix setup using CXCL2 208 as a classical neutrophil chemokine. Neutrophil trajectory analysis in dense collagen gels 209 (4mg/ml) showed that absence of PHD2 did not affect neutrophil chemokine sensing because 210 their directionality towards CXCL2 remained unaltered ( Figure 2D) and a similar strong 211 increase in cell speed was found in both PHD2-deficient and WT neutrophils ( Figure 2E). 212 Thus, these results show that PHD2-deficient neutrophils display an enhanced migratory 213 capacity in dense 3D collagen gels and that PHD2 loss does not affect CXCL2-induced 214 chemotactic capacity. In other words, the faster migration of cKO P2 neutrophils is independent 215 of chemotaxis induction and is rather linked to enhanced undirected motility or chemokinesis. 216

217
The migratory capacity of several cell types in complex microenvironments is highly 218 dependent on their capacity to deform when encountering narrow pores. 37,38 Therefore, we 219 evaluated whether cKO P2 neutrophils can overcome severely constricted spaces of only 1µm 220 width ( Figure 2F). 17,39 Remarkably, PHD2-deficient neutrophils showed an enhanced 221 preference to pass through these constrictions ( Figure 2G) and were also faster compared to 222 WT neutrophils ( Figure 2H). Interestingly, under these conditions, cKO P2H2 neutrophils 223 showed reduced migration; and, similar migration kinetics than WT cells ( Figure 2I, J), again 224 suggesting a PHD2/HIF2α-dependent axis in migration through extreme narrow constrictions. 225 226 Next, we studied whether the ability of cKO P2 neutrophils to pass through small confinements 227 is related to changes in their deformability when an external force is applied. For this, we first 228 analyzed neutrophil deformability using real time fluorescence and deformability cytometry 229 (RT-FDC) (see supplemental data), which can extract the stiffness of cells (Young´s Modulus) 230 in high-throughput, without contact at ms-timescales. 40,41 We used steady-state BMDNs, 231 Phorbol 12-Myristate 13-Acetate (PMA)-activated BMDNs, and peripheral blood neutrophils 232 isolated at 6h after thioglycolate-induced peritonitis. However, no differences were observed 233 between cKO P2 and WT neutrophils under any of the conditions tested (supplemental Figure  234 2B). Likewise, a microcapillary microcirculation mimetic (MMM) assay 42,43 using peritonitis 235 neutrophils showed no difference in their ability to passively navigate through multiple 236 constrictions at high speed (supplemental Figure 2C). Taken together, these assays strongly 237 suggest that loss of PHD2 does not affect neutrophil deformability under externally applied 238 stress without confinement. 239 11 240

PHD2-deficient neutrophils extravasate faster in vivo and accumulate in inflamed tissue 241
Based on the enhanced ability of PHD2-deficient neutrophils to overcome very small 242 constrictions, we decided to study the behavior of these cells in vivo; specifically, in a more 243 complex setting of sterile skin inflammation. Ear lobes from cKO P2 and WT littermate mice 244 that displayed no difference in total numbers of hematopoietic stem cells, myeloid progenitors 245 or mature neutrophils (supplemental Figure 3A), were ectopically treated with PMA and the 246 recruitment of Ly6G + cells was visualized using intra-vital 2-photon microscopy ( Figure 3A). 247 In line with our previous experiments, we found that PHD2-deficient neutrophils were able to 248 extravasate about 30% faster from the vessel into the ear tissue than their WT counterparts 249 ( Figure 3B-C). Furthermore, the cumulative effect of faster neutrophil extravasation time 250 resulted in an anticipated increase in Gr1 + cells in the inflamed cKO P2 ear compared to that 251 in WT littermates at 24 hours after PMA-treatment ( Figure 3D). Conversely but consistently, 252 this difference in migration was abolished in cKO P2H2 mice ( Figure 3E), further confirming 253 a role for HIF2α activity in driving increased migration capacity of these neutrophils. 254 As previous studies have described PHD2-related improved survival of neutrophils during 255 inflammation, 4,44 we evaluated the level of apoptotic cells in 24 hour PMA-treated ears, but 256 found no difference in cleaved caspase-3 + cell numbers (cCas3 + ) between the different 257 genotypes ( Figure 3F, G; supplemental Figure 3B, C). Additionally, as recent work has 258 associated PHD2 with enhanced neutrophil glycolysis and their recruitment to sites of 259 inflammation, 36 we assessed the glycolytic capacity of BMDNs from cKO P2, cKO P2H2, and 260 their respective WT counterparts by measuring extracellular acidification rate (ECAR). In line 261 with previous reports, PHD2-deficient neutrophils appeared to be significantly more glycolytic 262 than their respective WT counterparts ( Figure 3H). However, neutrophils lacking both PHD2 12 and HIF2α also showed significantly higher glycolysis ( Figure 3I). Taken together, although 264 HIF2α directly controls the migration speed of neutrophils in confined spaces and inflamed 265 tissues, this effect is independent of their survival or glycolytic activity. 266 267

HIF2α stabilization upon loss of PHD2 affects cytoskeletal gene expression profiles 268
It is well-accepted that the functionality of innate immune cells varies depending on the lipid-269 type composition of its cytoplasmic membrane. 45,46 Therefore, we evaluated if altered 270 membrane lipid composition of the cKO P2 neutrophils could account for their different 271 migratory ability, by performing high-throughput lipidomic analysis of freshly isolated 272 BMDNs (see supplemental data). However, as there were no significant alterations between 273 the cKO P2 and WT BMDNs (supplemental Figure 2D and Table 3), it appears unlikely that 274 differences in the lipid composition are directly responsible for the dramatic difference in the 275 migratory capacity of the cKOP2 neutrophils. 276 277 Next, to further characterize the molecular underpinnings of the HIF2α-driven neutrophil 278 migration phenotype, we used next generation sequencing (NGS) wherein the steady state 279 transcriptome of BMDNs derived from cKO P2 and cKO P2H2 mice were analyzed and 280 compared to that from their respective WT counterparts ( Figure 4A). Gene signatures of 281 various lineages were evaluated using gene set enrichment analyses (GSEA) as described 282 previously. [47][48][49] In line with our in vivo cCas3+ results, we detected no significant apoptosis 283 signatures among any of the genotypes ( Figure 4B) and NGS confirmed a significant 284 enrichment of glycolysis/gluconeogenesis related genes in both cKO P2 and cKO P2H2 285 BMDNs ( Figure 4C). Strikingly, steady state BMDNs lacking PHD2, with or without HIF2α, 286 displayed a significant reduction in genes related to the innate immune response but not the 287 13 chemokine signaling pathway ( Figure 4D). Together, these observations suggest that 288 significant HIF2α-independent changes in glycolytic capacity and immune response of PHD2-289 deficient neutrophils can be likely linked to HIF1α activity, as previously suggested. 2,36 290 Conversely, a number of HIF2α-dependent gene signatures associated with PHD2 deficiency 291 related to function and structure of the neutrophil cytoskeleton, including Rho GTPase activity 292 ( Figure 4E). Additionally, using an integrative method, we identified a number of HIF2α-293 associated master regulators that could potentially control cellular cytoskeletal rearrangements 294 through transcriptional or protein regulation (supplemental Figure 4A). 295 296

PHD2-deficient neutrophils 298
Small Rho GTPases (RhoA, cdc42 and Rac) are the final molecular effectors that steer 299 cytoskeletal dynamics. In line with this, we identified numerous potential associations (direct 300 and/or indirect) among 49 genes/proteins and with RhoA and/or Cdc42 (bold lines), but not 301 Rac GTPase ( Figure 5A). Notably, 7 of these genes have been previously identified as being 302 associated with HIF2α binding sites (supplemental Figure 4B). 50 303 To substantiate this link between PHD2/HIF2α and Rho GTPases, we used an ex vivo 304 enzymatic assay to quantify the activity of these Rho GTPases in untreated freshly-isolated 305 BMDNs from cKO P2 and P2H2 mice. Interestingly, cKO P2 neutrophils exhibited diminished 306 RhoA and Cdc42 GTPase activity ( Figure 5B, C), while Rac GTPase activity was comparable 307 with WT neutrophils ( Figure 5D). Further, cKO P2H2 neutrophils displayed no significant 308 reduction in either RhoA, Cdc42 or Rac GTPase activity, suggesting that regulation of RhoA 309 and/or Cdc42 is dependent on the PHD2/HIF2α-axis ( Figure 5B-D). 310 Given this reduction in RhoA and Cdc42 GTPase activity in PHD2-deficient neutrophils, we 311 examined whether their direct inhibition in WT neutrophils can mimic the motility phenotype 312 14 displayed by cKO P2 neutrophils. Therefore, we performed a series of ex vivo 1D-migration 313 assays using the RhoA inhibitor CCG100602 (CCG) or the Rho inhibitor Exoenzyme C3 314 Transferase (C3) and found that while the use of low doses of CCG or C3 enhanced the speed 315 of migrating neutrophils in 3µm micro-channels ( Figure 5E), treatment of cells with a Cdc42 316 inhibitor (ML141) did not have any effect on the velocity of BMDNs ( Figure 5F). Taken 317 together, our data strongly argue for a PHD2/HIF2α-orchestrated regulatory loop in RhoA 318 GTPase activity-dependent motility of BMDNs. 319 320 The PHD2/HIF2α-axis controls neutrophil accumulation in joints during severe 321

inflammatory arthritis 322
To test the biological effects of the enhanced migratory capacity of PHD2-deficient 323 neutrophils, we subjected the different mouse strains to an autoantibody-induced inflammatory 324 arthritis model (K/BxN), which has been shown to be myeloid dependent ( Figure 6A). 51,52 cKO 325 P2 mice displayed enhanced swelling of the hind limbs compared to their WT littermates 326 ( Figure 6B) and this effect was sustained throughout the first 2 weeks of the experiment. In 327 line with our previous results, cKO P2H2 and their WT littermates displayed no difference in 328 swelling ( Figure 6C). To characterize the myeloid composition of the inflamed knee joints, we 329 performed flow cytometry analysis of the synovial fluid drawn on day 5, which revealed much 330 higher accumulation of neutrophils in cKO P2 knee joints (>3-fold increase versus WT), along 331 with slightly enhanced macrophages ( Figure 6D); immunofluorescence for Gr1 on knee joints 332 further confirmed this observation ( Figure 6E). Conversely, although no differences were 333 observed in joint swelling between cKO P2H2 mice and their WT littermates, their synovial 334 fluid showed a slight but significant reduction in neutrophil numbers at day 5 compared to cKO 335 P2 mice ( Figure 6F). Thus, also in arthritic joints PHD2/HIF2α is a central axis during the 336 initial stages of the inflammation. 337 Discussion 339 In the current work we have explored if hypoxia pathway proteins can directly regulate 340 neutrophil motility, and reveal that activation of HIF2α in mouse neutrophils due to constitutive 341 PHD2 loss enhances neutrophil migration through very confined environments independent of 342 chemotactic -, glycolytic-or apoptotic-activity. Using a combination of in vivo, ex vivo and 343 deep sequencing approaches, we provide evidence that these neutrophils have the capacity to 344 migrate faster than their WT counterparts, and that this phenotype may be directly related to 345 changes in their cytoskeleton mediated by a substantial reduction in RhoA GTPase activity. 346 Although it is generally accepted that neutrophils are the first immune cells to arrive in the 347 tissue during inflammation, the molecular basis of neutrophil recruitment, which encompasses 348 extravasation and interstitial migration, remain elusive. Further, neutrophil recruitment has 349 been evaluated using a variety of migration assays in multiple studies related to the innate 350 immune response, 53-56 including in the context of hypoxia pathway proteins, 2,4 but these studies 351 call into debate the role of adhesion molecules. 57,58 Here, we consistently show in 1D, 2D and 352 3D assays that neutrophils lacking PHD2 alone, and not both PHD2 and HIF2α, display 353 enhanced cell motility and that only in severely confined environments. This difference in 354 chemokinesis between cKO P2 and WT remained in a comparable setup using a chemokine as 355 attractant (chemotaxis), demonstrating that the enhanced migratory capacity regulated by the 356 PHD2-HIF2α axis is probably a cell intrinsic characteristic. 357 An important process during neutrophil recruitment is the final and time-limiting step of trans-358 endothelial migration (TEM), which is, in part, mediated by mechanical forces generated by 359 the migrating neutrophil itself. 59-61 We reveal a central role for HIF2α in this process. Indeed, 360 considering the narrow pores between neighboring endothelial cells during the early phase of 361 16 neutrophil diapedesis, 60 our results from multiple approaches reiterate two main observations, 362 viz., that greater numbers of cKO P2 neutrophils pass through small constrictions with 363 enhanced speed. Intuitively, these observations account for the shorter TEM-time in a local ear 364 inflammation model. The cumulative effects of such enhanced transmigration into inflamed 365 tissues that were observable even at later time points in two completely independent in vivo 366 models, i.e., inflammatory skin lesions and sterile arthritis. Indeed, it is possible that the 367 enormous increase in cKO P2 neutrophils is positively affected by the fact that once a pore is 368 opened, successive neutrophils are more likely to extravasate at this spot, enabling more 369 neutrophils to enter the interstitium (skin) or the synovium (joint) of the inflamed tissue. 60 370 Previous studies in a model of acute lung injury have reported enhanced glycolytic capacity of 371 PHD2-deficient neutrophils, potentially due to HIF1α stabilization, which also enhanced 372 neutrophil recruitment to the inflammatory site. 36 Here, we confirm enhanced glycolysis in 373 cKO P2 neutrophils and show that it is HIF2α-independent, strongly suggesting that glycolytic 374 metabolism does not underlie the chemokinesis phenotype described here. The absence of 375 differences in neutrophil apoptosis in vivo was corroborated by the RNAseq data from both 376 single and double knock-out neutrophils, implying that the prolonged inflammation phenotype 377 in cKO P2 mice was probably not due to persistence of the neutrophils. This is in contrast to 378 results obtained using in vitro approaches that describe reduced apoptosis in HIF2α over-379 expressing neutrophils, which then resulted in delayed resolution of the inflammatory 380 response. 4 This group also reported delayed apoptosis in PHD2-deficient neutrophils and 381 connected this to persistent inflammation. 36 We believe these discrepancies are related to 382 differences in the experimental models used. 383 Although several studies have linked the hypoxia pathway to the migratory capacity of a cell, 384 only a few have suggested a role for the PHD/HIF axis in regulating cell migration through 385 changes in cytoskeletal function. [20][21][22] In migrating neutrophils in vivo, dynamic polymerized 386 actin converges at the leading-edge of pseudopods, while stable actin with high acto-myosin 387 contractility assemble at the rear. Both polarization and maintenance of this cytoskeletal 388 asymmetry strongly rely on Rho GTPase activity. 62,63 Using deep sequencing data from 389 neutrophils of single and double transgenic lines, we show that a vast number of genes 390 associated with Rho GTPase signaling are either directly or indirectly regulated by HIF2α. 391 Interestingly, cKO P2 neutrophils displayed a significant downregulation of RhoA GTPase and 392 we show this to be directly associated with enhanced motility because RhoA-inhibitor treated 393 WT neutrophils behaved similarly in confined environments. These findings are similar to 394 those reported earlier, i.e., increased flux of RhoA-deficient neutrophils and aggravated tissue 395 injury in LPS-induced acute lung injury. 64 A potential explanation is that the partial RhoA 396 inhibition would primarily decrease dynamic cell protrusions, known to restrict cell migration 397 by competing with stable actin cables at the cell rear. 18 Alternatively, the HIF2α axis could be 398 directly involved in the induction of cell contractility, which promotes neutrophil and DCs 399 migration under strong confinement. 17,34 However, further efforts are required to identify the 400 specific molecular mechanism. 401 In conclusion, our results demonstrate that HIF2α-activation, due to constitutive loss of PHD-402 2, enhances the motility of neutrophils in highly confined surroundings, also during 403 inflammation. Importantly, this phenotype is independent of chemotaxis signaling, glycolysis 404 or apoptosis. Mechanistically, it is the reduction of RhoA GTPase activity that enhances the 405 motility of PHD2 deficient neutrophils through very confined microenvironments. These 406 findings highlight the potential deleterious effects of sustained HIF2α activity and may have 407 important implications for the uncontrolled use of hypoxia mimetic agents that are currently 408 licensed or are in phase II and III clinical trials.