key: cord-0918521-69i7y3it authors: Wrobeln, Anna; Leu, Tristan; Jablonska, Jadwiga; Geisthoff, Urban; Lang, Stephan; Fandrey, Joachim; Droege, Freya title: Altered hypoxia inducible factor regulation in hereditary haemorrhagic telangiectasia date: 2022-04-07 journal: Sci Rep DOI: 10.1038/s41598-022-09759-9 sha: b4b391aaf0e1add772c4923d63d45c4f06b14804 doc_id: 918521 cord_uid: 69i7y3it Patients with hereditary haemorrhagic telangiectasia (HHT), also known as Rendu–Osler–Weber syndrome, suffer from the consequences of abnormal vessel structures. These structures can lead to haemorrhages or shunt effects in liver, lungs and brain. This inherited and rare disease is characterized by mutations affecting the transforming growth factor-β (TGF-β)/Bone Morphogenetic Protein (BMP) pathway that results in arteriovenous malformations and studies indicate an impaired immune response. The mechanism underlying this altered immune response in HHT patients is still unknown. TGF-β interacts with hypoxia inducible factors (HIF), which both orchestrate inflammatory and angiogenic processes. Therefore, we analysed the expression of HIF and related genes in whole blood samples from HHT patients. We could show significantly decreased expression of HIF-1α on the mRNA and protein level. However, commonly known upstream regulators of HIF-1α in inflammatory responses were not affected, whereas HIF-1α target genes were significantly downregulated. There was no correlation between HIF1A or HIF2A gene expression and the severity of HHT detected. Our results represent a rare case of HIF-1α downregulation in a human disease, which underlines the relevance of HIFs in HHT. The study indicates an interaction of the known mutation in HHT and the dysregulation of HIF-1α in HHT patients, which might contribute to the clinical phenotype. . Clinical characteristics of 66 patients with hereditary haemorrhagic telangiectasia. TAE teleangiectasia, FH family history, GI gastrointestinal bleedings, PAVM pulmonary arteriovenous malformations, HVM hepatic vascular malformations, CVM cerebral vascular malformations, ESS Epistaxis Severity Score with a scale from 1 = mild epistaxis to 10 = severe epistaxis. Data is shown in number of patients [(n) and (percentage)] or mean ± SD. www.nature.com/scientificreports/ Altered gene transcription of HIF and hypoxia related genes in whole blood of HHT patients. According to RT-PCR analysis, HIF1A gene expression was reduced by approximately 50% in blood cells of HHT patients compared to non-HHT controls (Fig. 1A) . In contrast, HIF2A gene expression was upregulated in HHT patients compared to non-HHT controls (Fig. 1B) . The HIF-1α target gene EGLN1 (encoding PHD2) (Fig. 1C) , which is part of the hypoxic response, was also reduced by 50% in HHT patients compared to non-HHT controls. To confirm the biological relevance of altered HIF-regulation, the expression of different inflammation-related genes was analysed. The expression of HIF-1α target gene IL6 was significantly decreased in HHT patients compared to non-HHT controls ( Fig. 2A) . TNFA gene expression (Fig. 2B ) and the gene expression of NFKB showed no significant difference between the analysed groups (Fig. 2C ). Gene transcription of angiogenesis related genes in whole blood of HHT patients is not affected. Further analysis was performed of HIF-regulated genes with biological relevance to angiogenesis. The expression of HIF-1 target gene ADM (Fig. 3A) , ANGPT2 (Fig. 3B) and VEGF (Fig. 3C ) was not affected, comparing HHT patients and non-HHT control individuals. Decreased HIF-1α protein in whole blood of HHT patients. To gain insight on the effects of downregulated HIF1A gene expression at the protein level, we assessed HIF-1α protein levels in blood cells of HHT patients and non-HHT control individuals by immunostaining. HIF-1α protein level was significantly reduced by about 50% in blood cells of HHT patients compared to non-HHT controls (Fig. 4) . www.nature.com/scientificreports/ Score. There was no significant correlation between HIF1A gene expression and the individual Epistaxis Severity Score of HHT patients (Fig. 5A) . Furthermore, the correlation of HHT patients' individual Epistaxis Severity Score and HIF2A gene expression showed no significant correlation (Fig. 5B) . HIF-1α is shown in red, ACTIN in green and the nuclei in blue. As negative control, staining with secondary antibody only is shown for non-HHT controls. Quantification of HIF-1α amount is shown in (B). HHT patients showed significantly less amounts of HIF-1α. Unpaired t-test was applied to determine significantly differences between the groups. Data are shown as mean ± SD, **p < 0.01. www.nature.com/scientificreports/ HHT patients suffer from vascular disorders, which lead to recurrent bleeding, as well as immunological abnormalities are indicated in few studies 1, 4, 5, 21 . Angiogenesis and inflammation are closely related processes and HIF is a key player in both of them. HIF activation in both hypoxia and inflammation regulates the transcription of several genes related to angiogenesis, metabolism, cell proliferation and apoptosis 17 . For the first time, this study provides evidence that HHT might alter expression of HIF and, potentially, of some HIF-regulated downstream genes. Recurrent bleeding cause lower haemoglobin levels (11.8 ± 2.5 g/dl, Table 1 ), which reduces the oxygen supply to the tissues resulting in hypoxia. Moreover, this hypoxia should be latent but global throughout the body. This phenomenon led us to assume that the amount of HIF-1α protein would have increased or-with respect to the mRNA level-at least not altered 22 . Unexpectedly, RNA expression analysis in whole blood revealed significantly decreased HIF-1α expression in HHT patients' leukocytes (Fig. 1A) . Even target genes related to cellular hypoxic response of HIF-1α showed decreased or unchanged expressions in HHT patients (Fig. 1C ) which is consistent with a reduction of HIF-1α also at the protein level. We succeeded in performing HIF-1α staining in whole blood cryosections of human subjects and found reduced amounts of HIF-1α protein in HHT-patients' leucocytes ( Fig. 4) . Obviously, analysis of whole blood RNA and protein expression of HIF-α in particular will provide data for leucocyte expression because they are the only nucleated cells in the blood. Gene expression in leucocytes from HHT patients is also of interest due to their impaired immune response. Thus, it was of interest to see inflammation related HIF-1α target gene IL6 was significantly downregulated ( Fig. 2A) . On the other hand, there was no change in the expression of TNFα as a general marker of acute inflammation in HHT patients (Fig. 2B ). This finding was corroborated by unaltered NFKB expression in HHT patients (Fig. 2C) . This, again, is of interest with respect to HIF-1α expression because NFKB is believed to act upstream of HIF-1α gene expression; based on our result, no evidence exists for a role of NFKB to mediate the reduced expression of HIF-1α at the transcript level in HHT 19, 23 . As mentioned above, many studies describe abnormalities in the immune response of HHT patients, to the extent that patients had milder symptoms during COVID infections or are less prone to tumour development compared to the non-diseased population 24, 25 . Both the development of symptoms in COVID-19 and the development of some tumours as an effect of chronic inflammation closely depend on the immune response, which in part is related to the increased expression of HIF target genes 26, 27 . Of note, a very recent work by Taniguchi-Ponciano et al. showed that critically ill COVID-19 patients had significantly increased HIF-1α expression in leukocytes 28 . Thus, reduced HIF-1α expression in immune cells may protect HHT patients from the exacerbating activation of the immune system, as it occurs in COVID-19 infections. How might reduced HIF-1α affect leukocyte function in HHT? Impaired migration of human HHT monocytes was found in a study by Han et al. 29 29 may therefore now be understood as an effect of reduced HIF-1α. This may have contributed to an impaired motility. This is in agreement with our previous studies, where we could demonstrate impaired motility of mouse macrophages with a conditional Hif1α knock out 32 . Of note, HIF2A expression showed the opposite effect from what we observed with HIF1A, namely an increased gene expression in HHT patients (Fig. 1B) . In mice, the conditional knockout of Hif2a in cells of the myeloid lineage resulted in increased recruitment of neutrophils to deeper layers of the colon during colitis 33 . One could speculate that HHT immune cells with increased HIF-2α might react in the opposite way and not be able to migrate into tissues and activate the common immune response. Obviously, we have to take into account www.nature.com/scientificreports/ that HIF-1α is expressed in all immune cell types, while HIF-2α is more selectively found in neutrophils and NK cells under hypoxic conditions as reviewed in Palazon et al. 19 . Thus, differential composition of immune cells and in particular the higher numbers of neutrophils and NK-cells in HHT patients may explain the increase in the overall higher leukocyte HIF2A expression 21 . Following several studies' descriptions of the influence of immune cells on angiogenesis and the vascular abnormalities in HHT patients, we analysed HIF target genes influencing the angiogenesis [34] [35] [36] [37] . Surprisingly, changes in angiogenesis associated genes were not detected in HHT patients' leukocytes (Fig. 3 ). It appears that, at the very least, blood leucocytes do not contribute to the vascular phenotype in HHT by expressing and secreting significantly higher levels of angiogenic molecules. This stands in contrast to the report of increased plasma VEGF levels in HHT patients by Cirulli et al. and Sadick et al. but may simply indicate that tissues other than those that are immune secrete VEGF in this disease 38, 39 . As mentioned above, key molecules in HHT are components of the TGF-β/BMP-pathway 2 . Caused by the haploid deficiency of the TGF-β/BMP pathway, TGF-β levels are elevated in the plasma of HHT patients 39 . Following previous studies, this should lead to an increase in HIF-1α regulation [6] [7] [8] ; thus TGF-β itself is unlikely to be the reason for the reduction of HIF-1α in HHT patients. However, when the impaired TGF-β/BMP pathway occurring in HHT was stimulated by suppressing TGF-β signalling with TGF-β antagonists, rat kidney cells showed decreased HIF-1α expression 40 . We therefore assume that the full functionality of the TGF-β/BMP pathway is required for both HIF-1α expression and protein accumulation. To prove this hypothesis, further studies should aim to reconstitute TGF-β signalling in HHT leucocytes in order to increase HIF1A expression to normal levels. To evaluate the severity of the clinical presentation of the disease, clinicians are still searching for suitable biomarkers for HHT, which are missing in daily clinical use [41] [42] [43] . To that end, we correlated the results of HIF1A and HIF2A gene expression with the clinically determined Epistasis Severity Score of each individual HHT patient. Unfortunately, a significant correlation between disease severity and HIF gene expression was not detected (Fig. 5) which may indicate that the situation is more complex and cannot be fully explained by such a correlation. It was demonstrated that TGF-β/BMP pathway members and HIF-1α interact and form a multiprotein complex on the promoter of the TGF-β co-receptor (endoglin) and through that mechanism, influence vascular remodelling and angiogenesis 9 . This multi-protein complex might be disturbed in HHTs because of reduced HIF-1α. In addition to the known mutations in endoglin, HIF-1α could contribute and aggravate defects in TGF-β signalling leading to dysregulated angiogenesis in HHT patients. Taken together, our findings show that the regulation of HIF-1α in HHT patients does not support the canonical regulation during hypoxia or inflammation. The results point towards a downregulation caused by the haploid impairment of the TGF-β/BMP pathway, but the role of different other factors remain to be elucidated While reduced expression of the HIF1A-gene and HIF-1α protein has, to our knowledge, never been described in any disease, future studies might reveal the underlying mechanism. Selection of participants and clinical parameters. The study was approved by the ethics committee of the University Duisburg-Essen (20-9162-BO). Participants were informed and gave written consent in accordance with the Declaration of Helsinki. The study is registered at Clinical trials.gov (ID NCT04469517). Adult patients with confirmed HHT (fulfilled at least three out of four Curaçao Criteria and/or had a positive genetic testing) were included in this prospective study 1, 44 . Alongside patients' characteristics such as age and gender, factors that might also affect HIF production were documented. HIF affecting parameters include smoking habits, days spent at height (> 1000 m above sea level) and the median weekly time spent doing sports. The documented HHT related clinical parameters include frequency, duration and intensity of epistaxis, need for medical attention, transfusions and signs of anaemia (haemoglobin level), family history, mucocutaneous telangiectases and visceral arterio-venous malformations especially involving the liver, lungs, gastrointestinal tract and central nervous system. The recurrent nosebleeds were described using the Epistaxis Severity Score (ESS) 45 . In all cases, whole blood samples were taken from patients peripheral arm veins. All experiments were compared with blood from a non-HHT control group. The non-HHT control group exists of 26 volunteers (15 men, 11 women) recruited from university employees (mean age: 40 ± 13; min: 21, max: 63). The number of patients and non-HHT donors is listed in the figures for each experiment. RNA preparation and RT-PCR of whole blood. RNA from whole blood was isolated using the PAXgene ® blood RNA collection system (Qiagen, Mississauga, Canada) according to the manufacturer's protocol. 2.5 ml of whole blood was drawn from HHT patients or non-HHT controls into PAXgene ® blood RNA tubes (Qiagen, Mississauga, Canada) and stored at -20 °C. Complementary DNA (cDNA) was synthesised using one µg of RNA and Moloney murine leukemia virus reverse transcriptase (Promega, Walldorf, Germany) according to the manufacturer's instructions. Quantification of gene expression was performed by real-time PCR (RT-PCR) with SYBR green fluorescent dye (Eurogentec, Verviers, Belgium) and the CFX96™ Real Time System (Bio-Rad Laboratories GmbH, Munich, Germany). cDNA was amplified by 40 cycles of 95 °C for 15 s and 60 °C for 90 s with gene specific primers ( Table 2 ) and normalized to ACTB (actin). Primer specificity was checked by Primer-BLAST and confirmed by size analysis of the PCR amplicons 46 . Expression was calculated with the 2-ΔCT method for statistical analysis and set as an induction relative to non-HHT controls in figures. HIF-1α staining of whole blood. Following withdrawal, blood was immediately transferred in to PHEM buffer (3 mmol PIPES, 1 mmol/l HEPES, 0.5 mmol/l EGTA, 0.25 mmol/l MgCl 2 , 3.5 mmol/l KCl, pH = 7.2) con- sis. After elimination of outliers (ROUT-test), the groups were analysed for Gaussian distribution using the D' Agostino-Pearson test. An unpaired t-test was applied to Gaussian distributed data and Mann-Whitney U test to non-Gaussian distributed data. Statistical significance is displayed as *(p < 0.05), **(p < 0.01), ***(p < 0.001) or ****(p < 0.0001). Informed consent statement. Informed consent was obtained from all subjects involved in the study. As in many studies of rare diseases, we have to mention the limited size of the investigated patient cohort. Regarding gene expression only a small selection of HIF target genes was analysed, more genes could complement this. The datasets generated during and/or analysed during the current study are available from the corresponding author. Table 2 . Human primer sequences. Primer sequences of specific PCR products used for RNA quantifications of human whole blood via RT-PCR. ACTB TCA CCC ACA CTG TGC CCA TCT ACG A CAG CGG AAC CGC TCA TTG CCA ATG G ADM AGT CGT GGG AAG AGG GAA CT ATC CGG ACT GCT GTC TTC GG ANPT2 AAC TTT CGG AAG AGC ATG GAC CGA GTC ATC GTA TTC GAG CGG HIF-1α TCA CTG GGA CTA TTA GGC TCA GGT CTC CAT TAC CCA CCG CTG AA HIF-2α CGG AGG TGT TCT ATG AGC TGG ASCTT GTG TGT TCG CAG GAA IL-6 TGC ATC TAG ATT CTT TGC CTT TTT CCA CTC ACC TCT TCA International guidelines for the diagnosis and management of hereditary haemorrhagic telangiectasia Involvement of the TGF-β superfamily signalling pathway in hereditary haemorrhagic telangiectasia TGF-β activation and function in immunity Patients with hereditary hemorrhagic telangectasia (HHT) exhibit a deficit of polymorphonuclear cell and monocyte oxidative burst and phagocytosis: A possible correlation with altered adaptive immune responsiveness in HHT Immunological abnormalities associated with hereditary haemorrhagic telangiectasia Interdependence of HIF-1α and TGF-β/Smad3 signaling in normoxic and hypoxic renal epithelial cell collagen expression Hypoxia-inducible factor-2α and TGF-β signaling interact to promote normoxic glomerular fibrogenesis TGF-β/Smad3 activates mammalian target of rapamycin complex-1 to promote collagen production by increasing HIF-1α expression Synergistic cooperation between hypoxia and transforming growth factor-β pathways on human vascular endothelial growth factor gene expression Endoglin expression is regulated by transcriptional cooperation between the hypoxia and transforming growth factor-β pathways Regulating cellular oxygen sensing by hydroxylation Cold Spring Harbor Symposia on Quantitative Biology A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1α regulates the VEGF expression and is potentially involved in lung and vascular development Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells Widespread, hypoxia-inducible expression of HIF-2α in distinct cell populations of different organs The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis HIF-1, O 2 , and the 3 PHDs: How animal cells signal hypoxia to the nucleus Regulation of hypoxia-inducible factors during inflammation HIF transcription factors, inflammation, and immunity A cross-talk between hypoxia and TGF-β orchestrates erythropoietin gene regulation through SP1 and Smads Impaired release of neutrophil extracellular traps and anemia-associated T cell deficiency in hereditary hemorrhagic telangiectasia In vivo expression of mRNAs encoding hypoxia-inducible factor 1 HIF-1 expression in healing wounds: HIF-1α induction in primary inflammatory cells by TNF-α SARS-CoV-2 infection in hereditary hemorrhagic telangiectasia patients suggests less clinical impact than in the general population Cancer incidence in patients with hereditary hemorrhagic telangiectasia The ever expanding role of hif in tumor and stromal biology Hypoxia: A key feature of COVID-19 launching activation of HIF-1 and cytokine storm Increased expression of hypoxia-induced factor 1α mRNA and its related genes in myeloid blood cells from critically ill COVID-19 patients Reduction of endoglin receptor impairs mononuclear cell-migration HIF-1α deficiency perturbs T and B cell functions HIF-1α is essential for myeloid cell-mediated inflammation Hypoxia-inducible factor 1α is essential for macrophage-mediated erythroblast proliferation in acute friend retrovirus infection The importance of hypoxia-inducible factors (HIF-1 and HIF-2) for the pathophysiology of inflammatory bowel disease Influence of recombinant codon-optimized plasmid DNA encoding VEGF and FGF2 on co-induction of angiogenesis Bone marrow mononuclear cells activate angiogenesis via gap junction-mediated cell-cell interaction Effect of interleukin-34 on secretion of angiogenesis cytokines by peripheral blood mononuclear cells of rheumatoid arthritis Therapeutic mechanism of cord blood mononuclear cells via the IL-8-mediated angiogenic pathway in neonatal hypoxic-ischaemic brain injury Vascular endothelial growth factor serum levels are elevated in patients with hereditary hemorrhagic telangiectasia Patients with hereditary hemorrhagic telangiectasia have increased plasma levels of vascular endothelial growth factor and transforming growth factor-beta1 as well as high ALK1 tissue expression Mitochondrial pathology and glycolytic shift during proximal tubule atrophy after ischemic AKI Does the genotype of HHT patients with mutations of the ENG and ACVRL1 gene correlate to different expression levels of the angiogenic factor VEGF? Research on potential biomarkers in hereditary hemorrhagic telangiectasia Pentraxin 3 level is elevated in hereditary hemorrhagic telangiectasia and reflects the severity of disease-associated epistaxis Diagnostic criteria for hereditary hemorrhagic telangiectasia (Rendu-Osler-Weber syndrome) An epistaxis severity score for hereditary hemorrhagic telangiectasia PrimerBank: A resource of human and mouse PCR primer pairs for gene expression detection and quantification We are very grateful to Michael Denniss, for proof reading as a native speaker. Further, we want to acknowledge all blood donors. The authors declare no competing interests.