Biologic Agents in Crohn's Patients Reduce CD4+ T Cells Activation and Are Inversely Related to Treg Cells

Crohn's disease (CD) is a chronic inflammatory disease with a complex interface of broad factors. There are two main treatments for Chron's disease: biological therapy and nonbiological therapy. Biological agent therapy (e.g., anti-TNF) is the most frequently prescribed treatment; however, it is not universally accessible. In fact, in Brazil, many patients are only given the option of receiving nonbiological therapy. This approach prolongs the subsequent clinical relapse; however, this procedure could be implicated in the immune response and enhance disease severity. Our purpose was to assess the effects of different treatments on CD4+ T cells in a cohort of patients with Crohn's disease compared with healthy individuals. To examine the immune status in a Brazilian cohort, we analyzed CD4+ T cells, activation status, cytokine production, and Treg cells in blood of Crohn's patients. Patients that underwent biological therapy can recover the percentage of CD4+CD73+ T cells, decrease the CD4+ T cell activation/effector functions, and maintain the peripheral percentage of regulatory T cells. These results show that anti-TNF agents can improve CD4+ T cell subsets, thereby inducing Crohn's patients to relapse and remission rates.


Introduction
Crohn's disease (CD) is a chronic inflammatory disease of a complex interface with broad factors. CD is a dysfunction of environmental factors, genetic background, and microbiome dysregulation, resulting in an imbalance of the immune response. e first clinical manifestations in patients are weight loss, fatigue, chronic diarrhea, abdominal pain, and fever [1][2][3][4]. e incidence of CD is increasing worldwide, including in Brazil [5,6]. Several types of treatment, biological and nonbiological (anti-inflammatory, antibiotics, corticoids, and glucocorticoids), have been shown to treat CD's symptoms [7]. In Brazil, patients commonly receive nonbiological treatments, whereas the most commonly used biological agent is anti-TNF alpha, but it is not widely accessible. erefore, biological treatment is prescribed to patients with a higher disease risk to induce and maintain remission.
e immune response plays a pivotal role in the pathogenesis of Crohn's disease [8,9]. Although, the patient's immune response has not been fully characterized. Many cells play different roles in the gut immune response. T cells are described as tissue-resident in the gut and can contribute to tissue damage or recovery during disease development. T regulatory (Treg) cells are a specialized subset of CD4 + Tcells whose function depends on the transcription factor FOXP3 (forkhead box P3), essential for maintaining immune homeostasis [10]. Although, experimental models have shown that immune cells, such as Treg cells and CD8 T cells, can improve the immune response in colitis [11,12]. Furthermore, maintenance of immunological tolerance in the intestinal mucosa is also achieved by Treg cells, suppressing immune activation [12]. In addition, the expression of CD39, an ectonucleotidase with coexpressing CD73, characterizes a specific subset of CD4 + Treg cells with augmented suppressor function [13].
Given the potential of anti-TNF agents to improve the remission of the majority of CD patients, we decided to investigate the impact of anti-TNF agents on the CD4 + T conventional, Treg, and suppressor Treg cells. Since all these subsets are important to maintaining homeostasis in the gut, we hypothesize that anti-TNF agents can recover these subsets, decrease inflammation, and improve the immune response of CD patients. Patients that underwent biological therapy can recover the percentage of CD4 + CD73 + T cells, decrease CD4 + T cell activation/effector functions, and maintain the peripheral percentage of regulatory T cells.
ese results show that anti-TNF agents can improve CD4 + T cell subsets, thereby inducing Crohn's patients to relapse and remission rates.

Study Population.
Our study cohort included three distinct groups: healthy, biological therapy, and nonbiological therapy. e healthy group is comprised of 28 individuals. None of these subjects had ever been diagnosed with any other chronic or autoimmune disease. We included 41 adult patients who had previously been diagnosed with CD according to the usual endoscopic, histological, and imaging criteria, and disease activity was measured using Crohn's disease activity index (CDAI). Of these, a total of 25 patients (median age of 36 years old) were treated with anti-TNF agents and 16 with nonbiological therapy (median age of 41 years old) ( Table 1). All the subjects tested negative for HIV-1/2, HTLV-1/2, syphilis, HCV, and HBV. We also collected four tissue samples from patients who underwent surgery and a pathologist's report. e majority of the participants in the study were female. e Ethics Committee of the Hospital Israelita Albert Einstein approved this study. All samples were collected following informed written consent from all participants.

Immunofluorescent Staining of Colon Section.
Colon sections were cut 5 mm, and after heat-induced epitope retravel, cells were incubated overnight at 4°C with rabbit anti-human FoxP3 polyclonal antibody (Abcam). After washing, sections were stained with a secondary antibody anti-rabbit Alexa 568 ( ermoFisher) for two hours at room temperature. e sections were washed and incubated for two hours with the following antibodies: anti-human mouse CD4 (clone: RPA-T4), CD39 (clone: TU66), CD25 (clone: M-A251), and CD73 (clone: AD2) (BD Bioscience). Afterward, the sections were incubated with DAPI and mounted on slides using ProLong Gold antifade reagent ( ermo-Fisher). Images were acquired on the LSM 710 confocal microscope (Carl Zeiss) with a 20X objective. e colocalization was assessed using Zen 2012 SP2. e expression of CD4 + CD25 + FoxP3 + CD39 + CD73 + defined the Treg suppressor; for the Treg cells, the expression was used CD4 + CD25 + FoxP3 + . At least ten different fields were counted for each sample.

Statistics. Statistical analysis was performed with
GraphPad Prism version 7 (GraphPad Software, La Jolla, California, USA). e nonparametric Mann-Whitney test was used for all statistical comparisons and the Pearson correlation coefficient. All data were reported as medians with a 25-75% interquartile range (M; IQR); a p value of 0.05 or less was considered significant.

Clinical Characteristics and Laboratory Features.
Forty-one patients were treated with anti-TNF agents (21 with adalimumab and four with infliximab), and 16 patients treated with nonbiological therapy (Table 1) were analyzed. We also recruited 28 healthy subjects. ese included the majority, around 70% of females and approximately 40 years old (Table 1).

Anti-TNF erapy Increased CD4 + CD73 + T Cells in the
Periphery. In this cohort, the percentage of CD4 + T cells was elevated in both groups of patients (Figure 1(a)). T helper cells (CD4 + T cells) are essential to participate actively in the immune response and improve homeostasis in several diseases. Since we observe an increase of CD4 + T cells in the periphery, we ask if these cells will be related to activation markers. CD4 + CD73 + T cells show an increase in the group of patients who use the biological therapy compared with the other groups (Figure 1(b)). e expression of CD73 alone by CD4 + T cells contributes to adenosine, which helps to activate the profile of CD4 + CD73 + T cells. When we look at the fluorescence intensity for the CD4 + HLA-DR + marker, the biologic group had similar results to healthy subjects; however, the nonbiological group decreased the expression (Figure 1(c)). HLA-DR is important for T cell recognition and is used as a marker for T cell activation [14,15]. On the other hand, traditional activation markers such as CD4 + CD38 + , which are used as a prognostic marker in several cases, show an increase in patients undergoing nonbiological treatment compared with the two other groups (Figure 1(d)).

Biologic erapy Decreases CD4 + T Cells Phenotypic Characteristics of Memory and 17.
Since we observed a difference in activation markers expression, we investigated the CD45RO + and the transcription factor RORct + expression on CD4 + T cells. Our results show that biologic therapy can decrease the expression of these markers (CD4 + CD45RO + and CD4 + RORct + ) expression on CD4 + T cells, as shown in Figure 2. CD45RO is an important marker for the T cell memory profile, while RORct denotes the T cell subpopulation related to inflammatory responses, such as the secretion of IL-17, IL-22, and others [16,17].

Nonbiological erapy Influences IL-17/Treg Cell Ratio.
To test biological therapy's ability to interfere in the Treg cells' expression in Crohn's disease patients, we investigated the natural and suppressor Treg. Unfortunately, we could not observe the difference in the natural Treg cells (CD3 + CD4 + CD25 high CD127 − FOXP3 + ) in all groups (data not shown). However, when we analyzed the characteristics of suppressor Treg (CD3 + CD4 + CD25 high CD127 − FOXP3 + CD39 + CD73 + ), it detected an increase in the subpopulation of Crohn's patients with nonbiological therapy when compared with biological treatment (Figure 3(a)). is result suggests that nonbiological therapy could suppress the T cell immune response. Furthermore, in the healthy subjects, we could notice an increase in the expression of a subpopulation of Treg that has the CD39 marker absent in the analysis (CD3 + CD4 + CD25 high CD127 − FOXP3 + CD39 − CD73 + ) when compared with Crohn's patients using biologic agents (Figure 3(b)). Deaglio et al. demonstrate that the CD4 + Tcells maintain the expression of FOXP3 and CD25; without CD39, the CD4 + T cells recognize the Treg cell phenotype [18].
Since we observed a higher activation status of CD4 + T cells, we decided to stimulate PBMC with PMA/ionomycin. Our results did not show any significant difference in the production of IL-17, TNF-α, or IFN-c (data not shown). However, since we detected a higher expression of RORct on CD4 + T cells, we further investigated the ratio of CD4 + T cells secreting IL-17 and natural Tregs. Crohn's patients on nonbiological therapy show an IL-17/Treg ratio higher than the median value for patients under biological therapy (Figure 3(c)). Natural Treg cells are inversely related to CD4 + T cell activation (as measured by HLA-DR + and CD38 + expression in CD4 + T cells) (Figures 3(d) and 3(e)).

Suppressor Treg on Gut Tissue. To address Treg cells'
presence on gut tissue, we performed a confocal analysis of four patients before surgery (Figure 4(a)). We observed cells expressing Treg suppressor markers on the lesion area (p � 0.0571) (Figure 4(b)). e limitation of these results was the reduced number of biopsies that we could have access to. However, we believe that it should be investigated in future studies because it will provide important information about Crohn's pathogenesis and treatment efficacy.

Discussion
CD4 + T cells are responsible for the immune response's critical functions, including the secretion of diverse types of cytokines. CD4 + T cells play a pivotal role in CD, as they regulate proinflammatory cytokine production, such as TNF-α, an essential mediator for IBD pathogenesis. It has been reported that Crohn's patients have increased levels of CD4 + T cells in their gut tissue, especially cells involved in memory immunity, such as CD4 + CD45RO + CD69 + [19], even though we observed increased CD4 + T cells in both treatment groups. Our data show that the percentage of cells expressing CD4 + and CD45RO + expression decreased in TNF-treated patients compared to healthy controls. e same profile was observed when median fluorescence intensity was measured. We believe that the discrepancy between results is due to Bishu and colleagues [19] using tissue samples, whereas we tested blood samples. Additionally, the lower levels of CD4 + CD45RO + cell percentage compared to increased CD4 + T cells in blood may be due to the increased number of circulating naïve cells and more differentiated recruitment of inflamed gut cells.
CD38 is a surface marker most prevalent on cells associated with regulatory activity, such as T regulatory cells and central memory T cells [20]. We have little information about CD4 + CD38 + cells in IBD, but Song and colleagues [21] demonstrated an association of such cells with HIV proliferation and persistence in patients receiving antiretroviral therapy. e authors correlate the expression of CD38 on central memory T cells with increased proliferation and survival, thus rendering cells reservoir of HIV. Furthermore, we observed an increase in CD4 + CD38 + cells in corticoidtreated patients compared to healthy controls, while TNFtreated patients showed decreased levels. ese two distinct profiles suggest that despite corticoids being capable of inducing remission, they do not increase T cells' capacity to respond to newer antigenic stimulation [17,22,23]. At the same time, anti-TNF-α can reduce central memory T cells. Furthermore, vitamin D blood levels are known to be correlated with increased expression of CD38-expressing CD4 Tcells [24]. Nevertheless, it is noteworthy that IBD patients suffer from vitamin D deficiency [25]; thus, the two distinct profiles observed may be influenced by such vitamin deficiency.
Since we had alterations in the CD4 T cells and the activation markers, the question of who is regulating the immune response in patients arises? Moreover, if Treg cells can control the dysregulated immune response? Our results demonstrate no difference in the Treg (CD4 + CD25 + FOXP3 + ) blood levels, showing similar results as Grundstrom and colleagues [26]. However, previous reports [27,28] showed the opposite results; IBD TNF-treated patients who responded to monoclonal antibody treatment had increased blood levels of Treg. is difference could be related to the methodology, and the cohort has colitis and Crohn's patients. e induction of CD39 expression constitutes a common immunoregulatory mechanism triggered by STAT3-activating cytokines in the innate immune system [29,30]. In addition, CD39 mediates the suppressive activity of mouse and human Treg cells, probably through the generation of adenosine [18,[31][32][33]. CD39 is responsible for scavenging proinflammatory molecules and producing immunosuppressive ones. Mice expressing higher proportions of human CD39 + and CD73 + cells showed attenuated symptoms from DSS-induced colitis [34]. Our results showed fewer CD4 + CD25 + CD73 + CD39 − FOXP3 + cells in TNF-treated patients than in healthy controls.
ese data indicate a shift in cell proportions due to the recruitment of anti-inflammatory cells to the inflamed gut. e confocal analysis confirmed the presence of Treg suppressor cells in the lesion area, suggesting that anti-TNF might induce an antiinflammatory profile. Furthermore, a higher 17/Treg cell ratio was associated with nonbiological treatment subjects. In addition, we detected a negative correlation between Treg cells and CD4 + T cells activation markers (HLA-DR and CD38). ese data suggest that the biological treatment could improve Treg function and the 17/Treg cell ratio associated with reducing chronic inflammation.

Conclusion
e treatment of Crohn's disease in developing countries like Brazil is a clinical challenge for clinicians and researchers. Understanding how biological therapies work versus conventional therapy is essential to improving the quality of life, prognosis, and decreasing disease relapse. Here, we were able to show the primary treatment used in Brazil for Crohn's patients in the context of CD4 + T cell immune response. Although our data do not bear directly on the efficiency of treatments, they indicate that patients undergoing anti-TNF therapy show an immune response similar to healthy individuals.

Data Availability
Due to ethical concerns, clinical and laboratory data cannot be made openly available and are available from the corresponding author upon request.