Severe mechanical injury (SI) is one of the main reasons behind children's disability and mortality [1, 2]. SI triggers decompensation of the body's life support systems as a result of combined effect of such damage factors as traumatic mechanical damage, blood loss and hypoxia. Mechanical damage is the initiating factor: it triggers the release of damageassociated molecular patterns (DAMPs), which, in turn, can disrupt the cellular immune response to exogenous antigens and pathogen-associated molecular patterns (PAMPs), thus promoting dysfunction of the immune system. Extracellular ATP (eATP) is one of the endogenous tissue DAMPs that trigger and regulate the immune response to damage [3]. In trauma cases, the level of eATP grows persistently in the injury [4, 5]. This compound is one of the main components of the purinergic system; being a strong pro-inflammatory signal, eATP plays an important part in the T cell functioning regulation. As a powerful damage-associated molecular pattern, eATP basically initiates inflammatory response through purinergic P2R receptors. At the same time, the end product of eATP degradation, extracellular adenosine, being an immunosuppressor, plays an important part in limiting that response. Extracellular adenosine functions through the A2A adenosine receptors, blocks the T cell receptor (TCR) signal by inhibiting phosphorylation of the zeta-associated protein 70 (ZAP-70) and activates the activating protein 1 (AP-1), thus decreasing the production of IL2, expression of CD25 and inhibiting the T cell proliferation. The levels of eATP and extracellular adenosine, as well as their biological effects, are strictly regulated by the catalytic effects of CD39 (E-NTPDasa1) and CD73 (Ecto5'NTasa), ectoenzymes expressed on the plasma membrane of immune cells. CD39 metabolizes ATP to ADP, pyrophosphate and AMP. CD73 ectonucleotidase degrades AMP into adenosine and phosphate. Thus, CD39 and CD73 exonucleotidase secure a balance between pro-inflammatory action of ATP and antiinflammatory action of adenosine in the focus of inflammation [6–9]. In case of a severe trauma, there is usually a period of prominent immunosuppression the pathogenesis of which is largely shaped by the decreasing level of T-lymphocytes. Absolute and relative number of T helper subpopulations is a significant marker in determining the severity of the pathological process and predicting its outcome [10–13]. Establishing the levels of expression of CD39 and CD73 exonucleotidase on various populations of circulating lymphocytes is of great clinical importance in the context of diagnosing and predicting the outcome of a wide range of diseases [14]. Therefore, the purpose of this study was to identify informative immunological criteria for the traumatic disease severity and outcome prognosis as applicable to children. The identification relies on the assessment of absolute and relative number of T lymphocyte subpopulations and the level of expression of CD39 and CD73 ectonucleotidase in T_reg and Th17 populations in severe mechanical trauma cases.

METHODS

The study involved 43 patients (28 boys (65.1%), 15 girls (34.9%); 116 observation sessions) with SI, treated at the Department of Anesthesiology and Resuscitation of the Research Institute of Emergency Pediatric Surgery and Traumatology, Moscow, in 2020–2021. We used the laboratory of the National Medical Research Center for Children's Health for laboratory studies, which were prescribed 1 to 5 times, depending on the length of stay of a given child at the Department of Anesthesiology and Resuscitation (DAR). The mean age of the children was 13.0 (6.0–15.0) years (Me (Q₂₅–Q₇₅)). The time options for laboratory studies were the first, third, fifth, seventh and 14^th days from the day of injury.

The control group was comprised of 41 apparently healthy children; all of them underwent medical examination at the National Medical Research Center for Children's Health. The children were comparable in age and sex: age — 12.41 (7.4–16.2) years; 26 boys (63.4%), 15 girls (36.6%).

Assessing the injury, we relied on the Injury Severity Score (ISS), the Glasgow Coma Scale (GCS) and its modification for patients under 2 years old, the pediatric GCS (pGCS) [15].

The outcome of an SI was assessed with the help of the Glasgow Outcome Scale (GOS) and the Severe Injury Outcomes Scale (OISS) [16], which suggest the following categories: category 1 — full recovery (the patient can live and be as active as before the injury); category 2 — good recovery (there are consequences that do not limit the level of social adaptation, but the patient cannot return to the preinjury level of functional activity and needs further treatment or rehabilitation); category 3 — moderate disability (there are consequences that disallow return to the pre-injury functional level, but the patient retains independent living skills); category 4 — severe disability (the patient needs assistance of others and cannot live independently); category 5 — death. These scales were applied to assess the condition of the patient at discharge from the hospital.

The patients were included in the study if they had an SI (ISS ≥ 16) and were treated in the ICU. Concomitant acute inflammatory and chronic diseases were a reason for exclusion.

At the first stage, we analyzed the results from the control group and the SI group. At the second stage, we analyzed the two groups formed with the help of GOS and OISS, the favorable outcome group (SIfav, n = 24) and the unfavorable outcome group (SIunfav, n = 17) (tab. 1). The distribution into these groups was based on the scores: patients were allocated to the SIfav group if they scored 4–5 points on the GOS scale and 1–2 points on the OISS scale, and to the SIunfav group if they scored 2–3 points on the GOS scale and 3-4 points on the OISS scale. A group of fatal cases (SIdeath, n = 2) was described separately (tab. 1).  

We assessed the quantity of Th17 lymphocytes (Th17 — CD3⁺CD4⁺CD161⁺), regulatory T lymphocytes (T_reg — CD4⁺CD127^lowCD25^high) in the patients and established the level of expression of purinergic signaling receptors on T_reg (CD39⁺T_reg — CD4⁺CD127^lowCD25^highCD39⁺ and CD73⁺T_reg — CD4⁺CD127^lowCD25^highCD73⁺) and Th17 lymphocytes (CD39⁺Th17 — CD3⁺CD4⁺CD161⁺CD39⁺ and CD39⁺Th17 — CD3⁺CD4⁺CD161⁺CD73⁺). Two-platform technology enabled assessment of the quantitative indicators of the subpopulation composition of peripheral blood T lymphocytes. The absolute number of lymphocytes was calculated with the help of a Sysmex XT-2000i hematology analyzer (Sysmex Corporation; Japan). The preparation of samples for cytofluorimetric analysis included incubation of 100 µl of whole blood with 10 µl of monoclonal antibodies tagged with fluorochromes for 20 min in a dark place. The erythrocytes were lysed with BD FACS™ Lysing Solution (BD Biosciences; USA); the duration of incubation therewith in the dark at room temperature did not exceed 10–12 minutes. The resulting samples were analyzed in a Novocyte flow cytometer (ACEA Biosciences; USA). The surface markers used to determine lymphocyte subpopulations were as follows: CD45, IgG1, IgG2a, CD3, CD4, CD25, CD127, CD161, CD39, CD73 (Beckman Coulter, USA; BD Biosciences, USA; SONY corp., Japan).

We used MS Excel 2016 (Microsoft corp.; USA), Statistica 10 (StatSoft, Inc.; USA), and IBM SPSS Statistics 25 (IBM corp.; USA) to process the data obtained. The results are presented as a median (Me) and quartiles (Q₂₅–Q₇₅). Mann–Whitney U-test with Bonferroni correction enabled comparison of differences in the attributes. Spearman's rank correlation coefficient (R) was used to assess relations between the attributes. The significance of quantitative indicators was assessed and threshold values (cut-off points) chosen with the help of the receiver operating characteristic curve (ROC curve). The threshold values were determined factoring in the maximum sensitivity and specificity requirements. The conclusions were considered significant at p < 0.05 (*).

RESULTS

The analysis of data from the control and SI groups revealed a pronounced decline in the absolute number of T_reg and Th17 in the early post-traumatic period.  The values of these indicators in SI patients were significantly different from the respective values registered in the control group (tab. 2, tab. 3). Third to fifth day post injury, the T_reg/Th17 ratio was decreased in the SI group compared to the control group, which is the result of gradual growth of the level of Th17 from the third day on (tab. 2, tab. 3).

The dynamics of the absolute number of Treg and Th17 cells expressing CD39 and CD79 was similar to the dynamics of small subpopulations of CD4⁺ lymphocytes during the acute post-traumatic period, but the changes were more pronounced in case of T_reg cells (tab. 2, tab. 3).  The relative amount of CD39⁺T_reg in children with SI varied from 6.3 to 76.6% and significantly exceeded the value of CD39⁺Th17 (range of variability: 0.3–24.1%) (tab. 2). As for CD73, the relative number of this marker was significantly higher on Th17 (range of variability: 2.6–99.9%) than on T_reg (range of variability: 0.5–55.2%). We uncovered no significant differences with the control group.

However, at some observation sessions the registered values significantly exceeded the maximum levels seen in the control group (tab. 2).

The analysis of ectonucleotidase mean fluorescence intensity (FL) on T_reg and Th17 revealed differences for CD73 on T_reg. Compared to the control group, the FL values for CD73 three days after the injury were increased (tab. 2, tab. 3).

Correlation analysis revealed a relationship between the percentage of T_reg and Th17 expressing CD39 and CD73 and the intensity of marker fluorescence. In case of Th17, the percentage of enzyme-expressing cells increases slightly as the intensity of fluorescence of CD39 (r = 0.27; p = 0.002) and CD73 (r = 0.20; p = 0.018) grows (fig. 1A, B).  In case of Treg, as the fluorescence intensity grows, the share of enzymeexpressing cells increases for CD39 (r = 0.71; p < 0.001) and decreases slightly for CD73 (r = –0.18; p < 0.05; fig. 1). The strongest direct dependence was found for CD39⁺T_reg (fig. 1B).  

A comparative analysis of the post-traumatic period data from SIfav and SIunfav groups has shown a significant increase in the relative amount of Th17 that occurred first through third days in the SIunfav group. At the same time, there were no differences between groups in terms of the number of T_reg cells (tab. 4). The levels of expression of CD39 on T_reg and Th17 lymphocytes differed significantly in the SIfav and SIunfav groups: patients from the latter group had it considerably higher (tab. 4, fig. 2). We did not do the comparative analysis for the STMdeath group (n = 2) because of the small number of observation sessions (three), but it can be noted that in case of such patients, with the relative amounts of T_reg and Th17 being comparable, the registered expression of ectonucleotidase on T_reg and Th17 was extremely low (tab. 4).  

The following clinical cases show the dynamics of CD39 expression on T_reg and Th17 in patients with unfavorable (Case #1, fig. 3) and favorable (Case #2, fig. 3) injury outcome.  

The analysis of fluorescence of ectonucleotidase on T_reg and Th17 in children from the SIfav and SIunfav groups revealed that the respective parameter differed significantly between the groups in case of CD39 on T_reg. In the SIunfav group we registered a slight increase in the fluorescence of CD39 on T_reg days 1 through 7 post-injury (tab. 5). As for the SIdeath group, the fluorescence values there were as follows: FL CD39 T_reg — 3.95 (3.7–4.67), FL CD73 T_reg — 4 (2.55–4.55), FL CD39 Th17 — 6 .77 (5–8.55), FL CD73 Th17 3.52 (3.1–3.95). Compared to SIfav and SIunfav, the FL CD39 T_reg values there were extremely low (tab. 5). 

We built a ROC curve (both SIfav and SIunfav groups) for the indicators that proved to be of high prognostic significance in traumatic disease cases in children. A good quality division model was generated for CD39⁺ T_reg % (AUC = 0.741) and FL CD39 T_reg (AUC = 0.721). The resulting cut-off for CD39⁺T_reg was 44.4% (sensitivity — 66.6, specificity — 84.7) and FL CD39 Treg — 8.25 c.u. (sensitivity — 87.5, specificity — 62.5).

DISCUSSION

This study shows that a severe mechanical trauma in children unbalances the T_reg/Th17 ratio in the early post-injury period, the imbalance translating into a slight shift towards Th17 while the absolute number of T_reg and Th17 cells decreases noticeably. These findings are consistent with the data published by other researchers [11–13, 17]. Among Treg and Th17, the absolute number of cells expressing CD39 and CD73 also proportionally decreases in the critical period of traumatic disease.

The analysis of cells expressing CD39 and CD73 ectonucleotidase in CD4⁺-lymphocyte populations in children with SI revealed that the highest expression of CD39 in the Treg population is up to 76.6%, that of CD73 in Th17 — up to 99.9%. In apparently healthy children, by contrast, the CD39 expression in the Treg population ranged from 19 to 49%, and that of CD73 by Th17 — from 7 to 35% [18].

We have established that depending on the traumatic disease outcome, the expression of ectonucleotidase in children going through the early post-injury period may be different. Compared to the patients for whom the outcome was favorable, children from the SIunfav group had the percentage of CD39 on T_reg and Th17 increasing and the intensity of CD39 fluorescence on T_reg growing on days first through seventh post-injury. A possible reason therefor is the role played by ectonucleotidases, especially CD39, in enhancing the hydrolysis of eATP and accumulation of extracellular adenosine in the injury locus, which triggers a cascade of reactions through the A2R system of adenosine receptors, this cascade ultimately leading to suppression of the immune response to prevent massive tissue damage [14]. The direct correlation between the level of CD39 fluorescence and the percentage of CD39⁺T_reg that we discovered indicates that the proportion of Treg abundantly expressing CD39 ectonucleotidase increases in response to injury. Previous studies that involved healthy adult donors have shown that cells with a large amount of CD39 on T_reg hydrolyze ATP more efficiently [8]. As for the CD73 ectonucleotidase, we established no correlation between its percentage and fluorescence intensity, probably due to the fact that CD73 is found both on the cell membrane and in soluble form [19]. In the deceased patients, the identified values of ectonucleotidase expression were extremely low, which may signal of development of decompensation of the immune system functions when the injuries are extremely severe.

CONCLUSIONS

The results of the study indicate that in children, the expression of CD39 and CD73 in T_reg and Th17 populations is significantly associated with the severity of injury and may be used to predict outcome of the traumatic disease. A deeper understanding of the role of purinergic signaling in the pathogenesis of traumatic disease suggests therapeutic potential of biopreparations based on the soluble forms of ectonucleotidase that may be designed to manipulate the immune system in such critical conditions as severe traumatic injury [20].