T Lymphocytes in Patients With Nijmegen Breakage Syndrome Demonstrate Features of Exhaustion and Senescence in Flow Cytometric Evaluation of Maturation Pathway

Patients with Nijmegen Breakage Syndrome (NBS) suffer from recurrent infections due to humoral and cellular immune deficiency. Despite low number of T lymphocytes and their maturation defect, the clinical manifestations of cell-mediated deficiency are not as severe as in case of patients with other types of combined immune deficiencies and similar T cell lymphopenia. In this study, multicolor flow cytometry was used for evaluation of peripheral T lymphocyte maturation according to the currently known differentiation pathway, in 46 patients with genetically confirmed NBS and 46 sex and age-matched controls. Evaluation of differential expression of CD27, CD31, CD45RA, CD95, and CD197 revealed existence of cell subsets so far not described in NBS patients. Although recent thymic emigrants and naïve T lymphocyte cell populations were significantly lower, the generation of antigen-primed T cells was similar or even greater in NBS patients than in healthy controls. Moreover, the senescent and exhausted T cell populations defined by expression of CD57, KLRG1, and PD1 were more numerous than in healthy people. Although this hypothesis needs further investigations, such properties might be related to an increased susceptibility to malignancy and milder clinical course than expected in view of T cell lymphopenia in patients with NBS.


INTRODUCTION
Nijmegen Breakage Syndrome (NBS) (MIM #251260) is a rare autosomal recessive disease belonging to a group of chromosomal instability disorders. The disease is caused by mutations in NBN gene (MIM #602667) encoding nibrin. The defect leads to defective response to DNA double strand break repair occurring both physiologically and in response to ionizing radiation and radical-producing agents (1)(2)(3)(4)(5)(6). The principal clinical manifestations of the syndrome include progressive microcephaly, dysmorphic facial features, mild growth retardation, mild-to-moderate intellectual disability, and an increased predisposition to malignancies (7)(8)(9)(10).
Due to humoral and cellular immune deficiency (11)(12)(13)(14) patients with NBS suffer from recurrent infections. Low concentration of serum immunoglobulins and/or inadequate specific antibody response (13), are caused by general B cell lymphopenia (11,(15)(16)(17) and/or lower frequency of switched memory B-cells (18). Severe impairment in T-cell dependent antigen response and features of defective cellular immunity have been attributed to T cell lymphopenia and defective T lymphocyte maturation (13,18,19).
This prospective study was initiated in attempt to describe peripheral T lymphocyte maturation profile in patients with NBS according to the currently known differentiation pathway (20).

PATIENTS AND METHODS
Peripheral EDTA-K2 anticoagulated blood samples were collected between November 2016 and December 2018 from 46 patients with common Slavic 657del5 mutation in nibrin (21), and from 46 healthy subjects, with the same female-to-male ratio as in the study group. Detailed clinical data were collected at the time of patient's (or healthy control's) visit in the outpatient department. None of the patients was treated for malignancy or demonstrated other features of lymphoproliferation at their enrollment into the study. In case of previous malignancy, the interval between initiation of the study and the end of treatment associated with remission was at least 2 years. All healthy controls have been sex and age-matched and met additional requirement of smallest possible deviation from the patient's age. They were also free from infections and have not been vaccinated recently.
Distribution of basic lymphocyte populations, including T, B, NK, as well as T helper and cytotoxic lymphocytes, were determined by flow cytometry using the lyse-no-wash approach and Multitest six-color cocktails of antibodies with Trucount tubes, to determine absolute cell counts of respective cell populations (Becton Dickinson, cat. no. 644611) ( Table 1). Antibody manufacturer's instructions were followed during the staining procedure. At least 15,000 events were acquired to BD FACSCanto II flow cytometer, with lymphocyte gate based on CD45 expression and side scatter characteristics. Lyse-no-wash settings for the FACS Canto Clinical software were used without any custom modification. Briefly, 50 µl aliquots of blood were incubated with optimally titered antibodies for 15 min in room temperature. The incubation was followed by erythrocyte lysis using 0.45 ml of BD FACSLysing Solution (Becton Dickinson, cat. no. 349202) diluted according to the manufacturer's instructions. Definition of basic lymphocyte subsets, i.e., T, B, NK, CD4, and CD8 T lymphocytes was performed according to standard procedures (22). Absolute numbers of individual cell subsets  , recent thymic  emigrants; TN, naïve T lymphocytes; TSCM, memory T lymphocytes with stemcell like properties; TCM, central memory T lymphocytes; TEM, effector memory  T lymphocytes; TEMRA, revertant terminal effector memory T lymphocytes  expressing CD45RA; L-TEMRA, low differentiated revertant terminal effector  memory T lymphocytes expressing CD45RA; H-TEMRA, high differentiated  revertant terminal effector memory T lymphocytes expressing CD45RA; TD,  terminally differentiated T lymphocytes. were calculated based on proportion of the respective cell subpopulation and absolute lymphocyte count (22).
Peripheral T lymphocyte maturation profile was analyzed according to the currently known differentiation pathway, using six-color cocktails of mouse fluorochrome-associated monoclonal antibodies specific for human receptors and differential expression of CD27, CD31, CD45RA, CCR7 (CD197), and CD95 (23) (details on monoclonal antibodies are presented in Table 1). Co-expression of CD31 and CD45RA was used to identify recent thymic emigrants among CD4 + lymphocytes and a population of naïve T CD8 + lymphocytes including also recently emigrating cells from the thymus. All remaining cell subsets have been identified both within CD4 + and CD8 + T lymphocyte populations. Naïve population (TN) has been identified as CD27 + CD45RA + CD197 + , memory T lymphocytes with stem cell-like properties (TSCM) as CD27 + CD45RA + CD95 + , central memory (TCM) as CD27 + CD45RA − CD197 + , effector memory (TEM) as CD27 + CD45RA − CD197 − , terminal effector memory expressing RA (TEMRA), as CD27 − CD45RA + CD95 + , including low-(L-TEMRA, CD27 + CD45RA + CD197 − and high-differentiated H-TEMRA, CD27 − CD45RA + CD197 − ), and late effector memory/terminally differentiated subsets (TD), as CD27 − CD45RA − CD197. We also analyzed features of immunosenescence manifested by expression of CD57 and KLRG1 (24), as well as features of exhaustion associated with expression of PD1 (CD279) (25,26). Briefly, 100 µl of peripheral blood samples were incubated with an adequate amount of antibodies for 15 min in darkness in room temperature. The sample was then lyzed with BD FACSLysing solution (Becton Dickinson), washed twice with wash buffer (PBS+0.1% sodium azide), and after suspending in wash buffer-acquired into appropriately calibrated BD FACS Canto II cytometer and analyzed with BD Facs Diva v.7 software. Gating strategy applied throughout the study is presented on Figure 1. The same approach was applied for both CD4 + and CD8 + T lymphocytes. Patient data were compared with results obtained in healthy controls collected during the study, and the differences were analyzed with Mann-Whitney and Fischer's exact tests.
The study was approved by the Bioethical Committee at the Children's Memorial Health Institute, Warsaw (Poland) and carried according to Helsinki Declaration. Written consent for participation was obtained from all patients older than 16 years, and parents or legal guardians in case of patients younger than 16 years.
PD1 as a marker of exhaustion was detected on a significantly higher proportion of both CD4 + and CD8 + lymphocytes in NBS patients than in healthy controls, with median CD4 + PD1 + lymphocytes composing 10.7 vs. 2.6% (p < 0.01) and CD8 + PD1 + composing 7.3 vs. 3.3% (p < 0.01). However, statistical differences in absolute counts between the study cohort and healthy controls were found only within CD4 + population (33

DISCUSSION
This study for the first time presents a detailed analysis of T lymphocytes and their subpopulations that reflect the development of T cells in periphery in a large cohort of patients with Nijmegen breakage syndrome. Thus, our observations extend and supplement earlier information on T cell development in this rare, but still underestimated disease.
Nibrin is known to participate in T cell development (48, 49) by affecting thymic output (50) and the V(D)J rearrangement process (6,51,52) resulting in increased proportions in T lymphocytes expressing TCRγδ receptor (19) and shift toward increased proportions of cells in mature development stages (14,19,53). The purpose of this prospective study was to provide more detailed description of the T lymphocyte maturation process in patients with Nijmegen Breakage Syndrome than available from previous reports (19,53). We also searched for differences in the process between healthy subjects and patients  with the DNA double-strand break repair defect caused by mutations in NBN gene. The distribution of relative and absolute counts of basic lymphocyte subsets, i.e., T, B, NK, CD4 + , and CD8 + T lymphocytes in the cohort under study did not differ from previous reports (7,14,15,19,50,53). It was found that the naïve T lymphocytes subpopulation in NBS patients was significantly smaller in comparison to healthy controls, both within CD4 + and CD8 + cells, as expected in view of the previously reported reduced expression of CD45RA (19,53). Our data demonstrated also, that thymic production of T helper lymphocytes measured by proportions and absolute counts of CD31 + CD45RA + cells was ineffective (50,53) and resulting in significantly lower number of naïve T CD4 + cells in NBS patients than in control subjects (Figure 3). The population of T CD8 + lymphocytes described by CD31 + CD45RA + immunophenotype was significantly more numerous than of naïve CD8+ cells (Figure 4). Even though CD31 + CD45RA + within T CD8 + are not limited to recent thymic emigrants (40), significantly lower proportions of CD31 + CD45RA + and TN indicate that thymic production of naïve T CD8 + lymphocytes is significantly affected by the mutated variant of nibrin.
Unexpectedly, TSCM were found to compose similar proportions of CD4 + and CD8 + , with significantly smaller absolute number of cells in NBS patients than in controls. Unaware of their existence, authors of previous reports included TSCM among naïve CD45RA + cells, as sharing common phenotypic characteristics (CD27 + CD45RA + CD45RO − CD197 + ) (19,53). Their functional properties are however completely different, as TSCM are antigen-experienced, and they exhibit effector activity in contrast to quiescent naïve T lymphocytes (36,54,55). Therefore, it seems that despite low thymic production, T lymphocytes in patients with NBS have enough potential to differentiate from naïve into more mature TSCM, but it is not sufficient to overcome low thymic production. Yet, this increased proliferative potential in comparison to physiology, might in turn potentially lead to the increased susceptibility to malignancies of lymphoid origin observed in NBS patients (9,10).
Homeostatic proliferation ensures the longevity of TCM in absence of cellular differentiation or activation. After proliferation, TCM can efficiently differentiate into effector cells (56). These processes seem not to be affected negatively by mutations in NBN, as TCM compose significantly greater proportion of both CD4 + and T CD8 + lymphocytes. TEM, which are potent effectors in healthy subjects, are also generated in greater proportions in NBS patients in comparison to healthy controls, but only within the CD4 + T lymphocyte population (Figure 3).
Similarly as in case of naïve cells, neither TCM, TEM, nor L-TEMRA within T CD4 + helper cells reached absolute counts observed in healthy controls. H-TEMRA and TEMRA composed similar populations, while TD CD4 + were even more numerous in NBS patients than in controls. Among CD8 + T lymphocytes, TEM, and L-TEMRA were significantly less numerous, while TCM, H-TEMRA, TEMRA, and TD reached similar absolute counts as in healthy controls (Figure 4). The explanation of important differences in maturation and proliferative potential between CD4+ and CD8+ lymphocytes reflected by the number of cells within individual populations requires more detailed molecular and functional analyses.
Preliminary data regarding age-related distribution of individual cell populations indicate that thymic output is deeply defective in almost all age groups. The defect seems to be more pronounced in younger children. In several patients, the absolute counts of all antigen primed CD4+ lymphocyte populations reached normal or almost normal counts. Adults appear to be an exception from this general observation, as recent thymic emigrants and naïve cells may reach low normal limits in some patients (Supplementary Figure 1A). This however does not mean that more cells are produced, as low limits of normal ranges in adults are lower than in young patients. Yet, more cells appear to be generated in several adults for antigen-primed populations. CD8+ T lymphocytes appear to behave differently in many terms (Supplementary Figure 1B). Deep defect in thymic production was observed in children and most adolescents, but not in adults. Children below 2 years of age seem to generate normal cell counts since reaching H-TEMRA maturation stage, while several patients from other age groups produce almost normal or normal counts of earlier antigen-primed cell populations. Moreover, several adolescents and adults seem to generate even higher than normal cell counts from the analyzed cell populations. These observations and their clinical context need to be verified in relation to larger group of healthy controls, when age related normal values will be established (study in progress).
Results of our study demonstrate that evaluation of the T lymphocyte maturation process in NBS patients by differential expression of only CD45RA/RO isoforms is misleading. Despite unsatisfactory thymic production, the generation of effector cells seemed quite effective and probably explaining relatively low incidence of clinical manifestations associated with cellular immune deficiency (17). Despite differences in biological properties, both TCM and TEM were correctly included in previous reports within the increased CD45RO + population (19). The same applied to the terminally differentiated populations, which were generated in significantly greater proportions in NBS patients in comparison to controls. TEMRA, which develop in parallel with TD (57), represent a revertant population that re-express CD45RA, but also express CD45RO isoform. They must have been incorrectly included in both naïve and memory populations (19,53). Therefore, previous reports overestimated both the population of naïve cells defined as CD45RA + , and memory population of CD45RO + cells (19,53). Mutations in the NBN gene are known to be associated with the telomere-initiated senescence (58,59). We analyzed the features of pre-term senescence of T lymphocytes in NBS patients by evaluating the expression of CD57 and KLRG1. Although CD57 + T lymphocytes are known to demonstrate cytotoxic abilities (49,60,61), the expression of CD57 is also associated with proliferative instability, correlating directly with the number of cell divisions and inversely with telomere length (43,62). We found significantly increased proportions of CD57 + T lymphocytes both among CD4 + and CD8 + populations ( Figure 5). This, however, could not be correlated (or solely correlated) to increased proportions of cytotoxic cells raised during viral infections, as no difference in CD57 expression between patients demonstrating chronic EBV viremia or EBV-free were observed (study in progress).
Surface expression of an inhibitory killer-cell lectin-like receptor G1 (KLRG1) identifies T lymphocytes that have undergone a large number of cell divisions (63). In healthy subjects, predominant expression of KLRG1 is observed on TEM and TEMRA cells (64). Both populations demonstrate potent effector functions, but are unable to proliferate (65). We found significantly increased proportions of KLRG1 + cells, corresponding to significantly increased proportions of TEM and TEMRA lymphocytes in patients from the study group (Figure 3). Similar suggestion regarding preterm T lymphocyte senescence in NBS was previously made by Meijerset al., based on results of studies performed in a significantly smaller group of patients (50). Therefore, considering the reported association of CD57 and KLRG1 with immune senescence (49,66) and results of our flow cytometric experiments we conclude that patients from the study group demonstrate features of preterm senescence.
Functional impairment of T lymphocytes termed "exhaustion" is associated with an increased expression of PD1 (24,66). We have observed significantly higher proportions of PD1 + T lymphocytes in NBS patients, both within T CD4 + and CD8 + subpopulations, which is in line with increased proportions of lymphocytes at the terminal differentiation stage. To our knowledge this feature has not been studied yet. Considering the features of an excessive proliferative history reflected by increased proportions of terminally differentiated and KLRG1 + T lymphocytes and increased proportions of PD1 + cells found in the study group in comparison to healthy controls, we conclude that T lymphocytes from NBS patients demonstrate features of exhaustion.
The difference between absolute counts of CD4 + and CD8 + lymphocytes demonstrating CD57, KLRG1, and PD1 expression appears to correspond with the observed differences in distribution and cell counts of the studied cell subsets (24,67,68). Significance of this discrepancy requires further analysis in context of clinical data.
We are aware of the limitations of the study. All experiments have been performed by flow cytometry and did not include either functional or molecular studies. Moreover, all references to functional properties of individual cell populations are based on published data. Such approach resulted mainly from limitations of the available research material, as large proportion of patients included minors, among them several below 2 years of age. Although functional properties of cells expressing CD57 or KLRG1 have not been evaluated in this study, premature senescence and shortened telomeres have been demonstrated in vitro in cultured NBN-mutated cells (69)(70)(71). Therefore, we feel the conclusion regarding premature senescence in NBS is justified.
In summary, we found significant aberration in peripheral T lymphocyte maturation process in NBS patients in comparison to physiological process. Despite low thymic production, the identified aberrancies and functional properties of individual T lymphocyte subpopulations lead to generation of significantly larger populations of effector T cells in NBS patients than in healthy people. Although this hypothesis needs further investigation, such properties might be related to an increased susceptibility to malignancy and milder clinical course than expected in view of T cell lymphopenia in patients with NBS.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Bioethical Committee at the Children's Memorial Health Institute, Warsaw, Poland. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

AUTHOR CONTRIBUTIONS
BP designed and supervised the flow cytometry experiments, analyzed and interpreted the data, and wrote the draft. BW-K designed the study, and collected and reviewed medical data. EH-P and AW collected and reviewed medical data. KT and UG performed the flow cytometry experiments. HG supervised the study. All authors critically revised and commented on the manuscript.