Transcription factors gene expression in chronic rhinosinusitis with and without nasal polyps

Twenty-two newly diagnosed, adult CRS patients (14 CRSwNP, 8 CRSsNP) were prospectively included at their first visit at the Department of Otorhinolaryngology and Head & Neck Surgery, University Medical Centre, Ljubljana, Slovenia. CRS diagnosis was established by symptoms, nasal endoscopy and a 2 or 3 mm computed tomography (CT) scan of paranasal sinuses. Included patients have never before used intranasal or systemic steroids. We took a biopsy of the middle meatal nasal polyp in CRSwNP patients or uncinate process mucosa in CRSsNP patients. Detailed patient inclusion data are shown in the flowchart in Figure 1 and detailed demographic data are summarized in Table 1. This study was conducted in accordance with the amended Declaration of Helsinki. The study was approved by the Slovenian National Medical Ethics Committee (approval number 34/10/12) and patients gave their written informed consent.

After sampling, the biopsy specimen was immediately fixed in 10% neutral-buffered formalin. Each paraffin-embedded specimen was sectioned at 5 micrometers thickness and stained by hematoxylin-eosin. The histopathologic evaluation of nasal biopsy specimens was performed using a structured CRS inflammation report as described by.² On hematoxylin-eosin-stained sections, the specimens were analyzed for eosinophil count per high

power field (HPF) (< 5, 5–10, > 10), neutrophilic infiltration, basement membrane thickening (≥ 7.5μm), subepithelial edema (absent to mild, moderate to severe) and presence of hyperplastic / papillary change, squamous metaplasia and fibrosis.

After sampling, tissue biopsy was stored in RNAlater solution (Qiagen, Hilden, Germany) at –40ºC. Total RNA was extracted using miRNeasy Mini Kit (Qiagen), according to the manufacturer’s instructions, and reverse transcribed to cDNA using the High-capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). RT-qPCR was performed on ABI PRISM 7500 Fast Real-Time PCR System at standard conditions utilizing TaqMan Fast Advanced Master Mix (Applied Biosystems). The Taqman assays T-bet (TBX21) (Hs00203436_m1), GATA3 (Hs00231122_m1), RORC (Hs01076122_m1) and FOXP3 (Hs00203958_m1) were utilized to determine the mRNA expression levels of transcription factors and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 4333764F) as an endogenous control (all Applied Biosystems). All measurements were performed in triplicate for each sample and relative expression was analyzed using the ΔΔCt method.¹⁹ Briefly, the relative expression of each mRNA was calculated by subtracting the Ct value of GAPDH from the Ct value of the gene analyzed to calculate the ΔCt.¹⁹ Then the relative expression of the sample was calculated using

Figure 1

Figure 2

Figure 3

the formula RQ sample = 2^{−(ΔCt sample−ΔCt calibrator)}. Nasal mucosa from the uncinate process of a control subject, 55 years old healthy female without CRS was used as a calibrator.

The data generated in the study were analyzed using GraphPad Prism 6.0 (San Diego, CA, USA). Statistical analysis was performed using the Mann-Whitney and Chi-square tests as appropriate with 2-tailed p. The data are expressed as numbers (percentages) or medians (first quartile (Q1) – third quartile (Q3)). The significance level was set at a p value of 0.05.

Twenty-two newly diagnosed CRS patients (14 CRSwNP and 8 CRSsNP) were prospectively included in the study. They have not been treated yet and have never used intranasal or systemic steroids. Both groups were comparable in age, sex, smoking, allergy, asthma, chronic obstructive pulmonary disease (COPD) and severity of CRS symptoms on visual analog scale 0–10. As expected, patients with CRSwNP had higher endoscopic Kennedy Lund scores than patients with CRSsNP. Exact clinical data are summarized in Table 1.

In concordance with CRS phenotypes, patients with CRSwNP had higher CT Lund MacKay scores at inclusion than patients with CRSsNP. Exact CT scores are summarized in Table 1. Representative CT scans of one CRSwNP and one CRSsNP patient are shown (Figure 2, 3).

Tissue eosinophilia > 10 per HPF was significantly more frequent (71%) in CRSwNP patients compared to CRSsNP patients (13%); meanwhile, there was no difference in tissue neutrophil infiltration between groups. As expected, moderate to severe subepithelial oedema was more pronounced in CRSwNP patients compared to CRSsNP patients. 10 CRSwNP patients with > 10 tissue eosinophils per HPF represented the eosinophilic CRSwNP group, the other 4 CRSwNP patients with < 10 tissue eosinophils per HPF represented the noneosinophilic CRSwNP group. More eosinophilic CRSwNP patients had basement membrane thickening ≥ 7.5μm compared to noneosinophilic CRSwNP patients. There was no difference in hyperplastic or papillary change, squamous metaplasia and fibrosis between groups. Exact histopathological data are summarized in Table 1.

GATA3 gene expression level was significantly upregulated in eosinophilic CRSwNP compared to noneosinophilic CRSwNP patients (Table 2 and Figure 4B). There was no difference in T-bet, RORC and FOXP3 expression levels between eosinophilic and noneosinophilic CRSwNP patients (Figures 4A, C, D). Effector Th cell master transcription factors T-bet, GATA3 and RORC gene expression levels in nasal mucosal were significantly elevated in CRSsNP compared to CRSwNP patients (Table 2 and Figures 5A, B, C). On the other hand, Treg master transcription factor FOXP3 gene expression level was equalized in CRSwNP and CRSsNP patients (Figure 5D).

We confirmed previous findings that CRSwNP is characterized by mucosal eosinophilia and pronounced subepithelial oedema.² Additionally, basement membrane thickening ≥ 7.5μm as a sign of remodeling and more severe inflammation was more pronounced in eosinophilic CRSwNP patients with > 10 tissue eosinophils per HPF compared to noneosinophilic CRSwNP patients with < 10 tissue eosinophils per HPF, in concordance with previous reports.¹⁷ Tissue eosinophilia in CRSwNP mucosa is known to be associated with type 2 cytokines and inflammation.^9,17

Surprisingly, we found significantly upregulated T-bet gene expression level in CRSsNP patients compared to CRSwNP. In contrast, one earlier study has reported downregulated T-bet in healthy patients inferior turbinates mucosa and CRSsNP mucosa compared to CRSwNP mucosa⁶, mean- while, other studies haven’t found any difference in T-bet expression between CRSwNP and CRSsNP groups.^20,21 T-bet is the master transcription factor of Th1 cells and type 1 innate lymphoid cells (ILC1), it promotes their differentiation, activation and interferon-γ (IFN-γ) secretion.^22,23 In response to IFN-γ signaling, T-bet can either activate type 1 inflammation or modulate it in case of exaggerated type 1 response.²⁴ Additionally, in effector cytotoxic type CD8+ T cells, T-bet is highly expressed.²⁵ Furthermore, in many acute and chronic viral infections, T-bet expression in effector CD8+ cells correlates with infection clearance or improved infection control.26,27,28 In our recent study, we found significantly depleted Th1 and abundant, effector cytotoxic CD8+ cells in CRSsNP nasal mucosa compared to CRSwNP.¹³ We might speculate here that highly upregulated T-bet in CRSsNP patients could originate from expanded effector cytotoxic CD8+ cells in CRSsNP nasal mucosa.

In this analysis, significantly elevated GATA3 gene expression level in patients with eosinophilic CRSwNP compared to patients with noneosinophilic CRSwNP was found, in concordance with previous reports.^21,29 GATA3 is the master transcription factor of Th2 cells^30,31 and type 2 innate lymphoid cells (ILC2)^32,33, enriched in CRS mucosa.³ Obviously, upregulated GATA3 expression in the eosinophilic CRSwNP group is related to type 2 (Th2 and ILC-2) inflammation. However, activated effector CD4-CD8- double negative T (DN T) cells can also secrete type 2 cytokines.^34,35 Importantly, in our recent study, elevated inflammatory DN T cells in the nasal mucosa of patients with uncontrolled CRSwNP were found. The master transcription factor of DN T cells is currently unknown. The impact and contribution of DN T cells to the type 2 inflammation in CRSwNP mucosa needs further investigation.

Unexpectedly, GATA3 gene expression level was significantly higher in patients with CRSsNP compared to patients with CRSwNP, in line with one earlier rtPCR study³, meanwhile, in contrast to two immunohistochemic studies.^6,20 In healthy sinus mucosa, GATA3 expression level was upregulated compared to CRSwNP mucosa, whereas there was no difference in GATA3 between CRSsNP mucosa and healthy sinus mucosa.³ Other studies found downregulated GATA3 expression level in healthy patient’s inferior turbinates mucosa compared to CRSwNP mucosa^6,36; however, inferior turbinate mucosa is not the same as sinus mucosa. Interestingly, CD8+ T cells were previously shown to express GATA3 constitutively, upregulate GATA3 upon T cell receptor (TCR) activation and further increase it by IL-4 and IL-2 stimulation.³⁷ Moreover, GATA3 expressing and type 2 cytokines producing effector memory CD8+ T cells were confirmed to be elevated in asthma³⁸ and tuberculosis infection.³⁹ Similarly to high T-bet expression, we might speculate here, that upregulated GATA3 in CRSsNP group could originate from activated effector CD8+ T cells, that were present abundantly in CRSsNP nasal mucosa in our previous study.¹³

We found significantly higher RORC expression in patients with CRSsNP compared to patients with CRSwNP, in line with previous reports.^6,20 Master transcription factor RORC is expressed by Th17⁴⁰ and type 3 innate lymphoid cells (ILC3).⁴¹ In our previous flow cytometric study, we reported the association of Th17 cells with well-controlled CRSwNP.¹³ The Th17 response was earlier found to be present in the healthy nasal mucosa and therefore, the authors proposed the Th17 cells were a part of normal homeostatic nasal mucosa immune response.¹⁵ Similarly, we could speculate here that upregulated RORC expression level in patients with CRSsNP is part of homeostatic nasal mucosa immune response, that might be better preserved and more functional in CRSsNP mucosa compared to CRSwNP.

FOXP3 expression level was equalized in eosinophilic CRSwNP compared to noneosinophilic CRSwNP as well as in CRSwNP compared to CRSsNP; these observations suggest that regulatory T cells are not significantly involved in inflammatory bias between CRSwNP and CRSsNP patients. Previously, contradictory results were reported with either no difference in FOXP3 expression levels between CRSwNP and CRSsNP²⁰ or downregulated FOXP3 in CRSwNP compared to CRSsNP.^6,42 FOXP3 gene expression was found downregulated in CRSwNP patients compared to healthy patients inferior turbinate mucosa.^6,36 FOXP3 is the master transcription factor for Treg⁴³; in contrast to the other master transcription factors, it has not been identified in ILCs.²³ Interestingly, some FOXP3+ Treg cells can even co-express another master transcription factor (T-bet, GATA3 or RORC), however, at much lower levels than corresponding effector Th cells. This way, Tregs may co-localize with corresponding Th effector cells to control their activation²³; T-bet and FOXP3 co-expression specifically inhibits Th1 and CD8+ T cell activation⁴⁴, GATA3 and FOXP3 co-expression controls type 2 inflammation and also co-modulates Treg function.⁴⁵ Additionally, co-expression of master transcription factors in effector Th cells was noticed in various chronic inflammatory diseases and can play a beneficial or a detrimental role.46,47,48 Moreover, upon in vitro stimulation of CRSwNP mucosa with Staphylococcus aureus enterotoxin (SE), upregulation of Treg was reported; the authors concluded, that Treg cells might be unsuccessful in inflammation control by Th2 activation and Th1 supression.¹² In the present study, we haven’t found any difference in FOXP3 expression levels between eosinophilic and nonesosinophilic CRSwNP patients and therefore, we cannot make any speculations about SE presence and stimulation in the eosinophilic CRSwNP patients mucosa.

Importantly, transcription factor-targeted future treatment strategies might have a positive impact on CRS mucosa by blocking the inflammatory pathway and, on the other hand, a negative impact by targeting co-expression in Tregs and their corresponding inflammation suppression.

This study has some limitations. No control group was included. The groups of patients were relatively small. However, all participants were newly diagnosed and have not been treated with intranasal or systemic steroids yet. Finally, we have only analyzed master transcription factors gene expression levels and not protein expression levels. We plan to expand gene expression investigation in CRS mucosa further to RNA sequence analysis.

To conclude, the type-2 inflammation was confirmed by elevated GATA3 gene expression level in patients with eosinophilic CRSwNP in comparison to patients with noneosinophilic CRSwNP. The unexpectedly found, simultaneous upregulation of T-bet and GATA3 transcription factors gene expression levels in patients with CRSsNP compared to patients with CRSwNP is currently unexplained; however, it might originate from activated CD8+ cells in CRSsNP nasal mucosa, in concordance with our recent findings of expanded effector cytotoxic CD8+ cells in the mucosa of patients with CRSsNP.¹³ The RORC upregulation in CRSsNP could be part of normal homeostatic nasal mucosa immune response that might be better preserved and more functional in CRSsNP mucosa compared to CRSwNP. In the perspective of future transcription factor-targeted topical treatments development, further data on transcription factors expression rates in CRS phenotypes are needed.