Immune Cell Profiles and Cytokine Levels of Patients with Interstitial Cystitis/Bladder Pain Syndrome.

Aims: To quantify the number of immune cells in the bladder urothelium and concentrations of urinary cytokines in patients with Interstitial Cystitis/Bladder Pain Syndrome (IC/BPS). To identify differences in these measures in IC/BPS patients with Hunner's lesions (IC/BPS-HL) and without Hunner's lesions (IC/BPS-NHL). Methods: Bladder tissue biopsies were obtained from 48 patients with IC/BPS-HL and unaffected controls (UC) and stained with antibodies for various immune cell markers such as CD138, CD20 and CD56. Levels of cytokines (Interferon (IFN)-{gamma}, Interleukin (IL)-1{beta}, IL-2, IL-4, IL-6, IL-8, IL12P70, IL-13, and TNF-) were measured from normalized urine obtained from 18 IC/BPS-HL, 18 IC/BPS-NHL, and 4 UC. Results: Numbers of CD138+ plasma cells, CD20+ B cells, and CD3+ T cells were significantly increased (50 fold, 30 fold, and an almost 3 fold increase, respectively; p-values: 1.34E-06, 3.26E-04, and 2.52E-6) in the bladders of IC/BPS-HL patients compared to UC. Patients with IC/BPS-HL had significantly elevated urinary levels of IL-6 (p=0.0028) and TNF- (p=0.009) compared to patients with IC/BPS-NHL and UC. In contrast, IL-12p70 levels were significantly higher in the patients with IC/BPS-NHL than in HL patients (p=0.033). No significant difference in IL-12p70 levels were observed between IC/BPS-HL and UC. Conclusion: Different cytokines were elevated in the urine of IC/BPS patients with and without HL, suggesting differences in underlying disease processes. Elevated levels of CD138+, CD20+, and CD3+ cells in HL indicate B and T-cell involvement in lesion formation. Determining which cytokines and immunological pathways are present in IC/BPS-HL could elucidate the disease mechanism.


Introduction
Interstitial Cystitis/Bladder Pain Syndrome (IC/BPS) affects millions of individuals within the US. 1 It is defined by bladder pain/discomfort accompanied by urgent voiding symptoms in the absence of other identifiable pathology. Most afflicted individuals have some portion of their pain derived from extravesical sources. IC/BPS is often difficult to diagnosis because symptoms are similar to other disorders, such as overactive bladder, vulvodynia, endometriosis, and prostatitis. 2 Identification of targets for diagnostic tests and treatments that consistently and accurately predict disease status and/or provide insights into disease mechanisms is an ongoing challenge. Evidence for immunologic dysfunction in IC/BPS includes the observation that several autoimmune disorders are significantly more prevalent in patients with IC/BPS than in the general population. These disorders include: Sjögren's syndrome, which has been found in 7.2% of patients with IC/BPS as opposed 0.5% in the overall population 5 ; rheumatoid arthritis, which has been observed in 4-13% of patients with IC/BPS compared to 1.0% of the general population 6 ; and systemic lupus erythematosus, which is observed in 1.7% of patients with IC/BPS, as opposed to 0.05% in the general population. 7 Immune system involvement in IC/BPS is also supported by the elevated counts of immune cells found in the bladder of patients with IC/BPS. Mast cells, among other immune cell types, have been reported to be elevated in the submucosal and detrusor layers of the bladder wall of patients with IC/BPS-HL and IC/BPS-NHL. 8 Mast cells can produce and release numerous pro inflammatory mediators including Interleukin-6 (IL-6), Interleukin-8 (IL-8), prostaglandins, Histamine, and Vascular Endothelial Growth Factor (VEG-F). 9 Previous studies . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint have shown elevated levels of IL-6 and IL-8 in the urine of patients with IC/BPS, and have also linked urinary VEG-F levels with higher rates of IC/BPS bladder glomerulations. 10,11 Cells found within the bladder biopsies of patients with IC/BPS-HL have also been shown to exhibit expression of several B and T cell markers including cytotoxic T lymphocyteassociated protein (CTLA)-4, Cluster of Differentiation (CD) 20, and CD79A. 12 Other studies have also confirmed significant elevation in CD138 positive (CD138+) plasma cell counts within the bladder of patients with IC/BPS-HL. Substantial plasma cell inflammation (200+ cells/mm 2 ) was observed in 93% of IC/BPS-HL samples and 8 % of IC/BPS-NHL samples. 13 Another connection between IC/BPS and the immune system is the involvement of the Nerve Growth Factor (NGF) neurotrophin, which has been implicated in IC/BPS pathogenesis. 14 Multiple immune cells such as Mast cells and Macrophages can induce the increased production of NGF and its concentrations are elevated in the bladder wall of patients with IC/BPS compared to unaffected controls. 15,16 Urinary NGF levels, in turn, correlate with urinary urgency and pain severity among patients with IC/BPS, providing a potential mechanistic link between immune cells and disease symptoms. 14 The majority of studies in the current literature have focused either on soft tissue or urine analysis from generalized IC/BPS patient cohorts and have identified infiltrating immune cells in the bladder, as well as elevated concentrations of multiple cytokines and antibodies in the urine. 17 1, 18, 19 However, different studies have identified different cytokines as being upregulated in IC/BPS. 10,16,20 It is possible that variation in the conclusions from these studies is due to inconsistent disease classifications between treatment centers. In addition, while immune cell infiltrates, including C20+ and CD138+ B cells have been observed, a profile of the relative abundance of the different immune cell types within patients is still necessary to clarify the role of the immune system as it pertains to this disease. 12,13 Our study assessed both urinalysis and bladder histology to create bladder immune cell profiles of patients with IC/BPS-HL and to detect differences in urinary cytokine concentrations between IC/BPS-HL, IC/BPS-NHL, and unaffected control (UC) sub-populations. Average numbers of CD138+ and CD20+ B cells, and CD3+ T cells in IC/BPS-HL bladder walls were significantly elevated compared to unaffected controls (UC). Patients with IC/BPH-HL had significantly elevated urinary levels of Interleukin 6 and Tumor Necrosis Factor-α compared to patients with IC/BPS-NHL and UCs.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint

Immunohistochemistry staining
All work received prior approval from the Institutional Review Board at Northwell Health. The presence of HL was confirmed by cystoscopy prior to sample retrieval. Thin 4-6 µm tissue sections were cut and stained with Hematoxylin and Eosin (H&E). Using similar thin sections, immunohistochemical (IUC) stains were performed on each biopsy sample for CD3, CD20, CD14, CD15, CD56, and CD138, for the quantification of T cells, B cells, monocytes, eosinophils/neutrophils, Natural Killer (NK) cells, and plasma B cells respectively (Figure 2A).
Hematoxylin was used as the counterstain. All antibodies were obtained from Ventana medical systems, Tucson, AZ. Antibody product details are as follows: CD3 (Catalog #: 790-4341, Lot Based on the IUC staining, counting was performed and recorded per high-dry field (x400) in the area of maximum diffuse infiltration by the inflammatory cells. Areas of distinct lymphoid follicle or tight aggregate formation were not counted to prevent skewing of results.
However, the presence of such lymphoid-predominant areas was noted. Biopsies from negative controls were examined and quantification of cells was performed along with the study cohort.
Both staining and cell counting was conducted in a blinded manner. Multiple counts were recorded from each patient and the average cell count for each marker was used during data analysis. A 2-tailed independent samples T-test was utilized to assess the statistical significance of the results. UC research participants were tested in duplicate. Plate readings were conducted using an MSD Quickplex SQ 120 imager (Model no: 1250) and the data used for analysis was obtained using . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

Mesoscale Discovery Cytokine assay
The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint the MSD workbench software program (version 3.0.18). A 2-tailed independent samples T-test was utilized to assess the statistical significance of the results.

Creatinine Normalization
Urine from each patient was tested for Creatinine concentration in order to determine the appropriate volumes to be used in the aforementioned multiplex cytokine assay. Samples were run in duplicate and averaged to obtain the final result. An Invitrogen Creatinine urinary detection kit (Thermofisher catalog number: EIACUN) was used and the manufacturer protocol was followed.

Sample Collection
Biopsy samples were acquired from the Hunner's Lesions of 48 patients with IC/BPS-HL after obtaining informed consent. The presence of HL (similar to those seen in Figure 1) was confirmed by cystoscopy prior to sample retrieval. Biopsy samples from 2 patients with no morphologic or clinical features of IC or any other urological malignancy were also retrieved, stained, and used as unaffected controls (UC). Urine cytokines in the present study were detected using the mesoscale discovery (MSD) U-plex assay: a highly sensitive electrochemiluminescence plate assay. Nine cytokines were . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint selected for testing based on previous literature and role in the immune system: IFN-γ, TNF-α,   IL-1b, IL-2, IL-4, IL-6, IL-8, IL-12p70, and IL-13. IL-6 is a key cytokine involved in B cell activation 25

Discussion
This study indicates that there are many significant differences in both the number of immune cells within the bladder and cytokine concentrations in the urine between patients with IC/BPS-HL and unaffected controls.
One of the more striking differences in patients with IC/BPS-HL was the significant elevation in numbers of cells expressing the B lineage CD20 and CD138 cell markers (Figure   2A and B). This drastic increase in the proportion of CD20+ and CD138+ cells indicates that B and plasma cells may have a specific role to play in the development or pathology of Hunner's lesions. This observation can serve as a starting point for identifying candidate mechanisms, diagnostic markers, or therapeutics for IC/BPS-HL. 19 These results are consistent with previous findings that also observed significantly increased CD20+ and CD138+ cell counts within the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Significant differences between patients with the IC/BPS-HL and IC/BPS-NHL cohorts
were also observed in urinary TNF-α levels ( Figure 3B, HL: 4.551 pg/mL, NHL: 3.084 pg/mL, p value: 0.009). TNF-α is produced by numerous immune cells such as mast cells, monocytes and macrophages and is a central mediator of the immune response. 26 The increased urinary TNF-α concentrations found amongst the IC/BPS-HL cohort builds upon previous findings which describe significantly elevated levels of TNF-α in the sera of patients with IC/BPS and identify the cytokine as a possible mediator of bladder urothelium apoptosis, which is observed in IC/BPS. 30,31 As TNF-α is increased in overactive bladder as well as IC/BPS, 26 measurements of TNF-α concentration on their own will be unlikely to produce a stand-alone clinical test for IC/BPS. However, the consistent observation of increased TNF-α in patients with IC/BPS will be useful for establishing immunological models that could ultimately lead to improvements in diagnosis and treatment. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint

Conclusion
Our results indicate that activation of B-cells and plasma cells correlates with IC/BPS-HL disease status. This conclusion is based on the high number of CD138+ and CD20+ cells found in the bladder tissue and the significant increase in IL-6 urinary concentrations among the HL cohort. While the present study has identified several immune cells present in the HL, the study design cannot determine whether the presence of these cells is a consequence or a cause of disease. It is likely that the increase in immune cell counts and coinciding upregulation of certain pro inflammatory cytokines represent a disease process in IC/BPS, but we cannot exclude that their presence may be neutral or even protective. Future studies should address the relationship between immune cell activation and IC/BPS disease severity in order to establish a causal relationship.
We did not find a statistically significant correlation between the concentration of cytokines in the urine of patients and the severity of clinical symptoms such as urinary urgency or severity of pain. The interpretation of this result is complicated because we cannot be certain whether it is due to natural variability in the levels of cytokines between different people or a true lack of a relationship between cytokine levels and disease. Longitudinal analysis of the relationship between clinical symptoms and degree of inflammatory changes can help resolve this issue.
It is still unclear how chemokines and cytokines in urine or serum correlate to bladder immune infiltration. Since chemokines and cytokines are indicators of inflammation, exploring the relationship between them and bladder immune cells in IC/BPS should be a high priority for future research. In the present study, biopsy samples were collected asynchronously with urine samples, so a correlation analysis could not be performed. While it is not always technically or ethically possible to collect urine and biopsy specimens from a given patient simultaneously, it is a direction that should strongly be considered in future studies.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint  The relative abundance of cells displaying the CD3 T cell marker fell from 52% among the unaffected controls to 29% among the IC/BPS-HL cohort.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint  Patients with IC/BPS-HL, compared against their age and sex matched IC/BPS-NHL counterparts, displayed significantly higher average urinary concentrations of IL-6 (HL: 3.23 pg/mL, NHL: 1.61 pg/mL, P value = 0.0175) and TNF-α (HL: 4.55 pg/mL NHL: 3.08 pg/mL, P value = 0.009). A significant difference in average urinary IL-6 concentrations was also observed between the IC/BPS-HL population and unaffected control (UC) population as well (HL: 3.23 pg/mL, UC: . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint 1.14 pg/mL, P value = 0.0028). The IC/BPS-HL cohort also displayed significantly lower urinary concentrations of IL-12p70 compared to the IC/BPS-NHL cohort (HL: 5.81 pg/mL, NHL: 8.91 pg/mL, P value = 0.0334). No other statistically significant differences in urine cytokine concentrations were observed among the 3 sub-groups. Urine from 18 IC/BPS-HL, 18 age-and sex-matched IC-BPS-NHL counterparts, and 4 control (without any inflammatory urological disease) research participants were tested in duplicate. The volume of urine used for each research subject was based upon the results of a prior Creatinine normalization assay. The data used for analysis was obtained using the MSD workbench software program (version 3.0.18).

Supplemental Material
Supplemental Table 1: Clinical characteristics and sex breakdown of the 3 research cohorts whose urine was collected for analysis with the Mesoscale discovery (MSD) multiplex cytokine assay. Proper informed consent was obtained from all patients prior to sample retrieval. (IPSS: International Prostate Symptom Score; UC: Unaffected controls; NHL: Patients with Interstitial Cystitis/Bladder Pain Syndrome without Hunner's Lesions; HL: Patients with Interstitial Cystitis/Bladder Pain Syndrome with Hunner's Lesions). Table 2: Clinical characteristics and sex breakdown of the 2 research cohorts who donated biopsies for immunohistochemistry analysis of bladder tissue samples. Proper informed consent was obtained from all patients prior to sample retrieval. Since UC patients did . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

Supplemental
The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint not present with urological symptoms, they were not asked about clinical information regarding their voiding and urinary associated pain, and therefor this information was not available in their clinical records. (UC: Unaffected controls; IC/BPS-HL: Patients with Interstitial Cystitis/Bladder Pain Syndrome with Hunner's Lesions) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted December 19, 2020. ; https://doi.org/10.1101/2020.12.17.20248414 doi: medRxiv preprint