Patients with high inammatory tendency induced by malignant stimulation through imbalance of CD28 and CTLA-4 / PD-1 leading to dopamine neuron injury

Parkinson's Disease is a common neurodegenerative disease in elderly. Parkinson's Disease patients were supposed to have a higher risk of malignant transformation, however, actually the incidence of various cancers in Parkinson's Disease patients is signicantly lower than common people.Parkinson's Disease patients are individuals with a high tendency of inammation, whose peripheral immune represented in an activated state. The secretion of inammatory factors in peripheral blood of Parkinson's Disease patients is signicantly increased. We hypothesized that the hyperinammatory predisposition of Parkinson's Disease patients is one of pathogenesis. Method: DBA/1 mice were used to simulate highly inammatory individuals, and the carcinogen DEN was treated for malignant stimulation.HE staining was used to observe the formation of lung tumors.Apoptosis of neurons was observed by TUNEL staining.Immunohistochemical and Flow cytometrywere used to observed the expression of CD4, CD28, MHCII, CTLA-4 and PD-1. IBA-1 + iNOS was used to label M1 type microglias, and IBA-1 + Arg1 was used to label M2 type microglias by immunouorescence. The contents of pro-inammatory cytokines TNF-α, IL-1β and anti-inammatory cytokines IL-10 and IL-4 in serum and brain tissues of mice in each group were detected by ELISA. malignant stimulation through imbalance of CD28 and CTLA-4 / PD-1 leading to dopamine neuron injury.


Introduction
Parkinson's Disease (PD) is a common neurodegenerative disease in elderly. The epidemiological studies of PD demonstrated that the occurrence of PD is contrary to the incidence of malignant tumors; The incidence of lung cancer [1], colon cancer [2] and other cancers in PD patients is signi cantly lower than that of common people [3], suggesting that individuals with PD tendency perform stronger cancer resistance.On the other hand, cancer is closely related to PD, some pathogenic genes of PD are also oncogenes. Parkin, PINK1 and SNCA are not only pathogenic genes of Parkinson's disease, but also closely related to the incidence of various malignant tumors such as ovarian cancer, lung cancer and liver cancer [4][5][6][7]. PD patients were supposed to have a higher risk of malignant transformation, however, actually the incidence of various cancers in PD patients is signi cantly lower than common people.
PD patients are individuals with a high tendency of in ammation, whose peripheral immune represented in an activated state. The secretion of in ammatory factors in peripheral blood of PD patients is signi cantly increased [8], the multiple in ammation-related receptors are over-expressed on peripheral immune cells, including antigen-presenting protein MHCII. These Phenomena suggest the abnormal activation of peripheral immunity in PD patients. Moreover, it has been found that many pathogenic proteinsof PD are actively involved in in ammatory pathways and homeostasis of in ammation [9]. In addition, as the genome sequencing of more than 1,000 PD patients, 24 gene loci were detected to be related with the risk of PD, among which 14 loci were con rmed to be involved in the regulation of immune in ammation [10].
We hypothesized that the hyperin ammatory predisposition of PD patients is one of pathogenesis.People with high in ammatory tendency under stimulation by external malignant change could produce excessive in ammatory response, preventing the occurrence of tumors, however causing the neurodegenerative damage in the brain.
In this study, DBA/1 mice were used to simulate highly in ammatory individuals, and the carcinogen diethylnitrosamine (DEN) was treated for malignant stimulation. DBA/1 is an inbred mouse and widely used in immune in ammation studies. DBA/1mice are spontaneously form of arthritis and also used to prepare the collagen induced arthritis model. DBA/1 mice are used to establish the in ammation model of central nervous system,having a high tendency of in ammation background. Compared with C57BL/6, the incidence of tumor in DBA/1 was reduced, however, the damage of dopaminergic neurons in the brain was signi cantly increased.Moreover, the T-cell-dominated in ammatory response was enhanced by the administration of CTLA-4 and PD-1 inhibitors, leading to a reduced incidence of tumor and enhanced death and neurodegeneration of dopamine neurons.This study demonstrated a close relationship among the high in ammation, tumor, and neurodegeneration, as well as regulating in ammation on the outcome of cancer and neurodegeneration. Hematoxylin-eosin (HE) staining HE staining was used to observe the formation of lung tumors.At week 4 and week 8, Blood samples were taken from the orbital vein and sera were immediately isolated.Thenmice in each group were sacri ced by dislocating their cervical spine. Following that, the lung tissue was collected, xed by paraformaldehyde. Tissue was dehydrated by ethanol at different concentrations, made transparent with xylene, embedded and block mounted in para n.Section were cut and stained with hematoxylin-eosin. An OLYMPUS BX-43 was used to capture images.

TUNEL
Apoptosis of neurons was observed by TUNEL staining.Para n-embedded sections of brain tissue were cut into 5 μm sections and treated with 50 ml TUNEL reaction solution. The sections were incubated at 37℃ for 60min, and then added with 50 μL streptomycin avidin-horseradish peroxidase solution incubating for 30min. The nuclei were stained with DAPI; After dehydration, the slices were sealed with an anti-uorescence quenching agent. The uorescence was detected under a uorescence microscope and photographed.

Immunohistochemical
Para n-embedded brain tissue wax blocks were sectioned at 5 μm. The citric acid antigen repair buffer was placed in the microwave oven for antigen repair, and the sections were placed in 3% hydrogen peroxide solution to block endogenous peroxidase.3%BSA was added for blocking at room temperature for 30min.Then anti-CD4, anti-CD28, anti-CTLA-4, anti-PD-1 and anti-MHCII antibodies were separately added to examinethe number of in ltrating T cells in the brain.The tissue sections were covered with HRP-labeled secondary antibody and incubated at room temperature for 50min.DBA kit was used for color development, hematoxylin was used to re-stain thenucleus.After gradient alcohol dehydration and sealing, the slices were observed under microscope.
Immuno uorescence Brain sections were subjected to depara nized, hydrated, antigen repaired (0.01 M citrate buffer, pH 6.0) and goat serum closed at 37℃ for 30 min. Serum was poured out, and TH antibody was added to determine the survival of dopamine neurons. IBA-1+ iNOS labeled M1type microglia cells and IBA-1+Arg1 labeled M2type microglia cells were incubated at 4℃ overnight.The sections were washed with PBS and added with uorescence-labeled secondary antibody. The sections were incubated at 37℃ in dark for 30min and then washed with PBS.DAPI staining nuclei were added and incubated at room temperature for 10 min. The slices were rinsed by PBS and sealed with anti-uorescence quenching. Photograph were taken by uorescence microscope.

ELISA assay
The contents of pro-in ammatory cytokines TNF-α, IL-1β and anti-in ammatory cytokines IL-10 and IL-4 in serum and brain tissues of mice in each group were detected by ELISA kit.The operation steps were strictly carried out according to the instructions of the kit.

Flow cytometry
Mice were narcotized, blood was collected from the abdominal aorta, and EDTA anticoagulant was added. Blood lymphocytes of mice were isolated by blood lymphocyte separation kit, the second layer of milky white lymphocytes was carefully moved to another tube.After centrifugation, the collected lymphocyte cells were re-suspended with PBS, and antibodies of CD4, CD28, CTLA-4, PD-1, and MHCII were added respectively for ow cytometry detection.

Statistical analysis
SPSS19.0 statistical software (IBM) was used for statistical analysis.The experimental data were expressed as mean ± standard deviation. The results of the experiments were analyzed using the χ2 test for ratio rate comparison and one-way analysis of variance for the other comparisons.P<0.05 was consideredtobe statistically signi cant.

Results
DBA/1 mice with high in ammatory tendency showed decreasing the tumor incidence andcontinuous increasing of peripheral in ammation After carcinogenic induction, the tumor incidence of DBA/1 mice was decreased signi cantly compared with the C57BL/6.HE staining showed that tumors were observed in mice in theboth of CD and DD group (Fig. 1a); After eight weeks treatment by DEN, the tumor incidence was about 50% in DD groups, and 100% in CD group. It suggested that the incidence of tumor was reduced in individual with high in ammatory tendency after malignant transformation.Moreover, after DEN induction, the proin ammatory factors TNF-α and IL-1βin peripheral blood of DBA/1 mice were also increased, the antiin ammatory factors IL-10 and IL-4 were persistently decreased.Interesting, in C57BL/6 mice, TNF-α and IL-1β were increased rstly and then decreased, IL-10 and IL-4 were decreased rstly and then increased (Fig. 1b).

DBA/1 mice with high in ammatory tendency showed promoting of intracranial in ammation and increasing the neurodegeneration under induction of malignant change
After carcinogenic induction, microglia cells in the brain of DBA/1mice were signi cantly activated.M1type microglia cells were dominant in DBA/1 mice, M2type microglia cells were relatively reduced. Moreover,M1type microglia cells were signi cantly increased in DD group compared with the control group. In this study, IBA-1 + iNOS was used to label M1type microglias, and IBA-1 + Arg1 was used to label M2type microglias by immuno uorescence, and the results showed that the activation of microglia in DBA/1 mice was more than that in C57BL/6 mice (Fig. 2a).The damage of dopamine neurons in DBA/1 mice was moresigni cant than that in C57BL/6 mice under DEN induction.TH staining showed that the number of dopamine neurons in DBA/1 mice was signi cantly reduced compared with that in C57BL/6 mice (Fig. 2b). Tunel staining assay demonstratedthat the apoptosis of substantia nigra neurons of DBA/1 mice were higher than that of C57BL/6 mice (Fig. 2c).
CD28 and CTLA-4/PD-1 reduced theT-cell-dominated in ammatory response, reduce intracerebral in ammatory response and protectedthe neurodegeneration Peripheral T cells were activated and intracranial in ammatory response was signi cantly increased in the DD group.The ow cytometry showed that CD4, CD28 and MHCII in peripheral blood lymphocytes of DD group were signi cantly increased, while CTLA-4 and PD-1 were signi cantly decreased (Fig. 3a).Th1mediated TNF-α and IL-1β levels in the DD group wascontinued to increase, however,Th2-mediated IL-10 and IL-4 levels was decreased continually (Fig. 3b).T cells activationin the DBA/1 group was signi cantly increased, and the expressions of CTLA-4 and PD-1 were gradually decreased.To detect T cell number in mice brain under DEN stimulation, the expression of CD4, CD28, MHCII, CTLA-4 and PD-1 weremeasuredbyimmunohistochemicalstaining (Fig. 3c). The resultsshowedthat the expressions of CD4, CD28 and MHCII in the brain wereincreasedgraduallywith time, while the expression of CTLA-4 and PD-1wereedecreasedovertime in DBA/1 micecomparedto C57BL/6 mice.These results suggest that activated T cells participate and worsen intracranial in ammation.

Combination of CTLA-4 and PD-1 blocker can over-activate T cells, worsen peripheral and intracranial in ammation, reduce the incidence of tumor
The combinationof CTLA-4 and PD-1 blockers can reduce the incidence of tumors. DBA/1 mice were treated with CTLA-4 and PD-1 blocker after injection of DEN, and the tumor incidenceat 8weekstreatmentwas reduced to50% compared with DD group, while the tumor incidence inC57BL/6 mice was about 20%.The combination of CTLA-4 and PD-1 blockers can promote the activation of T cells.The results showedthe level of CD4, CD28, MHCII in peripheral lymphocytes were signi cantly increased in DDCP group and CDCP groupcompared to thatin the DD and CD group, however, the content of CTLA 4, PD-1 were signi cantly reduced (Fig. 4a), the changes in the DDCP group were the most evident. Furthermore, Immunohistochemistry results demonstrated the similar results that the expression of CD4, CD28 and MHCII in mice brain tissues was increased, while the content of CTLA-4 and PD1 was decreased in theDDCP group and CDCP group compared to that in the DD and CD group (Fig. 4b).Furthermore,ELISA results showed that CTLA-4 and PD-1 blockers could activate the in ammatory response after malignant stimulation in DBA/1, with increased TNF-αverify and IL-1β levels and decreased IL-10 and IL-4 levels (Fig. 4c).
Combination of CTLA-4 and PD-1 blocker can cause damage to dopamine neurons, promote the occurrence of neurodegeneration Under the combination treatment of CTLA-4 and PD-1 blockers, brain microglia cells were activated and dopaminergic neurons were damaged. Immuno uorescence showed that CTLA-4 and PD-1 blockers activated microglia activity in DBA/1 under DEN stimulation, and M1type microglia was increased, M2type microglia was decreased (Fig. 5a).At the same time, compared with the DD group, the number of dopamine neurons in the DDCP group was decreased (Fig. 5b), and the apoptosis rate of substantia nigra neurons was increased (Fig. 5c).

Discussion
Parkinson's disease is a common neurodegenerative disease,however,its etiology and pathogenesis are still unclear. PD is considered to be the result of the interaction of multiple factors. In this study, we demonstrated that harmful stimulation induced the individuals with a high tendency of in ammationtoactivate the T cells immune response, increase the stimulating factor, decrease the inhibitory factor, leading to excessive in ammation, spreading to the brain, eventually leading to the damage of neurons. Because the co-inhibitory molecules CTLA-4 and PD-1 on T cells can down-regulate the immune response, weinhibited the expression of CTLA-4 and PD-1, T cells was activated. Although the occurrence of tumor was reduced, the in ammatory response and the degenerative injury of neurons were aggravated in the brain.The results suggested that the body with high in ammatory tendency is at high risk of PD, and the disorder of immune mechanism may be one of the importantfactors of PD.
In this study, under malignant stimulation the incidence of tumor in DBA/1 mice was lower than that in C57BL/6, additionally, the peripheral pro-in ammatory cytokines TNF-α and IL-1β were continuously increased, the anti-in ammatory cytokines IL-10 and IL-4 were decreased, and the degeneration and damage of neurons in the brain was signi cant.In C57BL/6 mice, TNF-α and IL-1β reached the highest levels at the fourth week after stimulation and then gradually decreased.Weconsidered thatafterfourweeks' stimulation, thein ammatory responsein wildmicewas initiatedto down-regulated. However, in DBA/1 mice the in ammatory response was gradually increased, suggesting that in ammation in high-in ammation mice was deregulated, eventually leading to apoptosis and damage of dopamine neurons.Clinically, the high levels of IL-1β, TNF-αhavebeenreported in the serum and cerebrospinal uid of PDpatients [11].This is consistent with our results,thus it is considered that in ammatory factors may also be involved in the occurrence of central in ammation leading to the damage of dopamine neurons.We considered that the tendency of high in ammation in the body is a double-edged sword, which not only inhibits the tumorigenesis, but also initiates the neuronal damage in the brain.The pathogenesis of PD is closely related to in ammatory response. Combined withthe signi cantly decreasing of tumor incidence in PD patients, we speculate that the partially cause of PD patients is based on high in ammation tendency in the body, the body was induced to in ammatory response as facing to the various malignant stimulation in environment, however the in ammation is out of control, leading tothe death of dopaminergic neurons in the brain.
In this study, under DEN induction, following with CTLA-4 and PD-1 inhibitors treatment, CD4, CD28 and MHCII in peripheral and intracranial of DBA/1 mice were signi cantly increased, CTLA-4 and PD1 were signi cantly decreased, and intracranial pro-in ammatory factors were signi cantly increased.These results suggested that malignant stimulation could over-activate the T cells in highly in ammatory mice and simultaneously participate in and worsen the intracranial in ammatory response, leading to the occurrence of neurodegenerative injury.T cell activation has always been the research spot of anti-tumor immunity and tumor vaccine research. And T cell activation is also closely related to PD.SaundersEt al. demonstrated that the imbalance between memory T cells and regulatory T cells was closely related to the severity of PD [12]. T cell over-activation is also involved in exacerbating in ammation in the brain.In the studies of PD patient brain [13]and PD animal models [14], a large number of CD 4+ and CD 8+ T cells were in ltrated. In addition, various treatment in activating T cell immunity to inhibit tumor growth [15] also cause immune encephalitis [16], Transverse myelitis [17]. These studies suggesting that the overactivated T cell immune system in peripheral blood can cause in ammation of the central nervous system.Therefore, we considered that highly in ammatory individuals activate T-cell immunity in order to combat malignant stimuli, triggering the strong peripheral in ammation, inhibiting tumor,however, promoting in ammatory response and neurodegeneration in the brain.
In this study, under DEN induction, following with CTLA-4 and PD-1 inhibitor treatment,the tumor incidence rate in DBA/1 mice was decreased, CD28 content in peripheral blood and brain was increased, and the content of CTLA-4 and PD1 was further decreased.The content of pro-in ammatory factors in peripheral blood and brain tissue was also increased, and the content of anti-in ammatory factors was decreased, therefore the damage and apoptosis of dopamine neurons in brain were more signi cant.It was suggested that the combined application of CTLA-4 and PD-1 inhibitor further promoted T cells activation, aggravated the damage of dopamine neurons, and signi cantly decreased the incidence of tumor.These results showed that CTLA-4 and PD-1 could regulate the process of T cell activation and resist tumorincidence and aggravate the dopamine neuron injury. The activation of T cells requires dual signals, the rst signal comes from the speci c binding of MHC-antigen complex of Antigen-presenting cells (APC) withT cell receptor (TCR), and the second signal comes from the binding of co-stimulating molecules (CD28) or co-inhibiting molecules (CTLA-4 and PD-1) on the surface of T cells with the ligands on the APC. CD28, a co-stimulatory molecule on the surface of T cells, can activate T cells and promote in ammation after stimulation.The co-inhibitory molecules CTLA-4 and PD-1 can inhibit the activation of in ammatory T cells. CD28, CTLA-4 and PD-1 can precisely regulate the transition of T cells to pro-in ammatory Th1 oranti-in ammatory Th2, and affecting the activation state of peripheral immunity. The inhibitor of CTLA-4 performed good treatment e cacyin rheumatoid arthritis [18], Type 1 diabetes [19],Kidney transplantation immune rejection [20].On the contrary, co-inhibitory molecules such as CTLA-4 and PD-1 are increasingly expressed in tumors, which inhibit the activation of T cells and weaken the immune surveillanceof tumors, becoming the most important mechanism for tumor-cell escape [21].This study shows that the combination of CTLA-4 and PD-1 inhibitors can activate T cells in highly in ammation-prone animals and promote excessive in ammation, and increase the transformation of microglia into M1type cells to aggravate the damage of dopamine neurons, promoting the development of PD, however, reducing tumor escape and signi cantly decreasing the incidence of tumor.
The above results demonstrated that individualswith high in ammatory tendency are the basis for the occurrence of Parkinson's disease. As exposed to harmful stimuli, in ammation is initiated to protect the body from the occurrence of malignant tumors. However, due to their high in ammatory tendency,excessive in ammation is caused and eventually dopamine neurons are injured.The harmful stimulus may be not only con ned to DEN, it may also include a variety of microbial infection, chemicals, all kinds of physical and chemical stimulation, etc. These harmful stimuluses induced the excessive in ammation in the individual with high tendency, accumulated gradually in a long period leading to the damage of the central nervous system.This study con rmed that bodywith a high in ammatory tendency are at high risk of excessive in ammatory response to malignant stimulation. In the process, CTLA-4 and PD-1 as important regulatory factors could reduce the in ammatory response.As the block of CTLA-4 and PD-1 together lead to deterioration of in ammation, increasing the damage of neurons in the brain. Although it could curb the occurrence of tumors, increase the risk of nerve degeneration. These nding con rmed thatthe in ammation tendency of genetic background is the pathogenesis of PD risk factors.

Conclusions
High in ammatory tendency induced by malignant stimulation through imbalance of CD28 and CTLA-4 / PD-1 leading to dopamine neuron injury.