Impaired T-cell differentiation in diabetic foot ulceration

Article metrics


Foot ulceration is one of the most debilitating complications associated with diabetes, but its cause remains poorly understood. Several studies have been undertaken to understand healing kinetics or find possible therapies to enhance healing. However, few studies have been directed at understanding the immunological alterations that could influence wound healing in diabetes. In this study, we analysed the T-cell receptor (TCR) repertoire diversity in TCR-αβ+ T cells. We also analysed the distribution and phenotype of T cells obtained from the peripheral blood of healthy controls and diabetic individuals with or without foot ulcers. Our results showed that diabetic individuals, especially those with foot ulcers, have a significantly lower naive T-cell number and a poorer TCR-Vβ repertoire diversity. We also showed that the reduced TCR-Vβ repertoire diversity in diabetic individuals was mainly owing to the accumulation of effector T cells, the major source of tumour necrosis factor-α production, which was even more pronounced in patients with acute foot ulceration. Moreover, the expression of several inflammatory chemokine receptors was significantly reduced in diabetic patients. In conclusion, effector T-cell accumulation and TCR repertoire diversity reduction appear to precede the development of foot ulcers. This finding may open new immunological therapeutic possibilities and provide a new prognostic tool in diabetic wound care.


The prevalence of diabetes is increasing at alarming rates worldwide and is greatly affected by the concurrent overweight and obesity epidemic caused by poor diet, lack of physical activity and genetic predisposition. Lower limb amputation, one of the major causes of diabetes-associated morbidity and mortality, is a consequence of diabetic foot ulceration (DFU).1 In contrast to normal wounds that progress through the phases of wound healing, the wounds in chronic diabetics become stalled in different phases, and healing does not occur in a synchronised fashion. This healing defect is due to diabetes-associated neuropathy, microangiopathy and immune function impairment.2 Because our understanding of how the immune response is impaired in diabetic patients is still incomplete, DFU treatments have mainly relied on lower limb amputation,3 with some occasional new strategies providing very limited results.4, 5, 6 Although prevention is essential for DFU care,7 its impact is also limited because we are still unable to predict the onset of DFU. Indeed, despite the identification of various risk factors for DFU,8, 9, 10 a reliable marker is urgently needed to identify the individuals at risk or in the early stages of DFU. To develop better therapeutic strategies or to better predict the appearance of DFU, we must first understand why the immunological response is compromised during foot infections in some diabetic individuals and how this defect leads to chronic wounds that fail to heal.

Diabetic patients present a systemic pro-inflammatory environment with increased levels of interleukin (IL)-1, tumour necrosis factor (TNF)-α and IL-6, and regulated on activation, normal T-cell expressed and secreted (RANTES).11, 12 This pro-inflammatory environment is not only present in the blood but also extends to various other tissues such as the skin,13 and significantly contributes to the microvascular and macrovascular complications associated with diabetes.14, 15, 16, 17 To elucidate the effect of insulin resistance on inflammation, a recent study using a mouse model of high-fat-diet-induced glucose intolerance revealed that insulin resistance, by impairing inflammation and collagen fibre organisation, increased wound healing time.18 In fact, hyperglycaemia alone was able to activate nuclear factor-κB,19 which impairs leukocyte activation20 and migration.21 Conversely, pro-inflammatory cytokines such as IL-6 and TNF-α were shown to regulate resistin expression in humans, worsening insulin resistance in diabetic patients.22 Despite the chronic inflammation observed in diabetic patients,11 the diabetic immune system is unable to mount the effective immune response required for bacterial control and tissue regeneration.23

Results from some DFU models also suggest that the impaired wound healing in diabetics would not only originate from an excess of pro-inflammatory cytokines but also from a deficit of anti-inflammatory and healing-associated cytokines like transforming growth factor-β and IL-10.24 Because these antagonistic cytokines usually inhibit each other’s action, the distinction between the cause and the effect is unclear. In fact, the fine balance between inflammation and pathogen tolerance is mainly regulated by T-cells. These cells also control the adaptive response through cell–cell interactions or through cytokines. In animal models, CD4+ T cells depletion negatively impacts wound healing, whereas CD8+ T-cell depletion has a positive impact.25 Despite the possible differences between the murine and human T-cell responses, a decrease in the number of CD4+ T cells was also observed in diabetic individuals.26

Various studies have demonstrated that infections that cause a strong T-cell stimulation may provoke a deleterious effect on the T-cell receptor (TCR) repertoire diversity, and therefore extensively affect the immune response to subsequent infections by different pathogens.27, 28, 29, 30, 31 This effect is mainly attributed to the competition for cytokines that mediate the survival of naive cells, such as IL-2, thereby imposing a limiting factor on the total T-cell count.32 Moreover, other studies have also shown that the accumulation of small effects caused by encounters with regular pathogens can result in a significant effect on the TCR repertoire diversity of aged individuals.33, 34 Nevertheless, using a transgenic murine model, it was also established that a fair amount of plasticity exists in the TCR repertoire, mainly because of the cross-reactivity phenomenon that can compensate for a reduction in diversity.35 However, this effect may not be sufficient to fully overcome the decrease in diversity, which likely limits the defence capacity against natural pathogens.

Therefore, the pro-inflammatory environment and the recurrent wound infections in diabetic patients may have an effect on the TCR repertoire diversity and on the distribution of T-cell populations. However, we still have to understand with precision how it affects the outcome of wound healing in diabetics. To address this question, we studied the TCR-αβ+ T-cell subpopulations present in the blood of diabetic patients with or without DFU to analyse their distribution and TCR repertoire diversity.



Peripheral blood (PB) samples from 22 healthy adult controls and 41 diabetic patients, including 21 without DFU, 10 with acute DFU (<3 months from diagnosis36) and 10 with chronic DFU (>3 months from diagnosis36), were collected using ethylenediamine tetraacetic acid (EDTA)-K3 tubes for immunophenotyping. The sample characterisation data and relevant clinical information are summarised in Table 1. In addition, larger PB samples from 6 healthy individuals, including 4 females and 2 males aged between 35 and 47 years (average 41 years), were collected into EDTA-K3 tubes for T-cell sorting and in vitro culture experiments. In addition, a small PB sample was collected into sodium heparin tubes for cytokine production assays.

Table 1 Sample characterisation

All diabetic patients and controls enrolled in the study had no previous history of neoplastic malignancies or autoimmune disorders and, at the time of sample collection, had no other clinically detectable infection than the foot infection in the case of DFU patients. A history of previous foot ulcerations was considered as an exclusion criterion for all groups except for the chronic DFU group.

TCR-β diversity assessment

Deoxyribonucleic acid (DNA) was extracted from the sorted populations using the Sigma GenElute Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich, St Louis, MO, USA) according to the manufacturer’s protocol. The extracted DNA was quantified using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).

Polymerase chain reaction (PCR) amplification was performed using 100 ng of DNA for each sample, as described in the Biomed II protocol,37 using the TCR-β Gene Clonality Assay (InVivo Scribe Technologies, San Diego, CA, USA). Briefly, three multiplex PCRs were performed, each amplifying different zones of the TCR-β locus. The first and second PCRs were designed for the detection of rearrangements between the Vβ and Jβ regions, including forward primers for the following Vβ families: Vβ2, Vβ4, Vβ5, Vβ6, Vβ7, Vβ8, Vβ9, Vβ10, Vβ11, Vβ12, Vβ13, Vβ14, Vβ15, Vβ16, Vβ17, Vβ18, Vβ19, Vβ20, Vβ21, Vβ22, Vβ23 and Vβ24. The reverse primers used in the first PCR targeted Jβ1.1, Jβ1.2, Jβ1.3, Jβ1.4, Jβ1.5, Jβ1.6, Jβ2.2, Jβ2.6 and Jβ2.7, and, for the second, Jβ2.1, Jβ2.3, Jβ2.4 and Jβ2.5. The Vβ primers covered ~90% of all the Vβ gene segments.

The third PCR was designed for the detection of rearrangements between the Dβ and Jβ regions, using forward primers for Dβ1 and Dβ2, and reverse primers for Jβ1.1, Jβ1.2, Jβ1.3, Jβ1.4, Jβ1.5, Jβ1.6, Jβ2.1, Jβ2.2, Jβ2.3, Jβ2.4, Jβ2.5, Jβ2.6 and Jβ2.7.

Amplification was performed using the phycoerythrin (PE) 9600 thermal cycler (Perkin Elmer, Applied Biosystems, Inc., Foster City, CA, USA), and product sizes were detected with the Applied Biosystems ABI 310 single-capillary electrophoresis system (Thermo Fisher Scientific) using a 47 cm × 50 μm capillary at the single-base sensitivity. The resulting data were analysed using the Peak Scanner Software v1.0 (Thermo Fisher Scientific).

T-cell immunophenotyping

The analysis of surface antigen expression on the PB T cells was consistently performed using a whole-blood direct immunofluorescence four-colour staining with the monoclonal antibodies (mAbs) indicated in Table 2.

Table 2 Monoclonal antibody specificities, clones and sources

The cells were prepared for flow cytometry as previously described.38 In brief, the blood cells were incubated for 15 min at room temperature with the antibodies after the erythrocyte lysis and cell fixation steps, which were performed using the FACS lysing solution (Becton Dickinson, East Rutherford, NJ, USA).

The flow cytometry experiments were performed on a FACSCalibur cytometer (Becton Dickinson) using the BD CellQuest software (Becton Dickinson) for sample acquisition and the BD Paint-A-Gate Pro software for data analysis (Becton Dickinson). Alternatively, we also used a Navios cytometer equipped with the Navios software for sample acquisition and the Kaluza software for data analysis (all from Beckman Coulter, Brea, CA, USA).

The different subpopulations were selected using anti-CD4 or anti-CD8 together with anti-CD27 and/or anti-CD28, and anti-CD45RO mAbs. This strategy, proposed by Appay et al.,39 allowed for a clear identification of three major blood T-cell populations: naive (CD27+ CD28+ CD45RO), activated/memory (CD27+ CD28+ CD45RO+) and effector (CD27 CD28).

Cell culture assays

Mononuclear cells were separated from whole blood by centrifugation using a 1.077 density gradient media (Lymphoprep, Axis Shield PoC, Oslo, Norway) and were cultured for 3 weeks at a density of ~1 × 106 cells per ml in 2 ml plates placed in a sterile environment at 37 °C under a 5% CO2 humidified atmosphere. Cells were cultured in complete medium (Roswell Park Memorial Institute medium (RPMI) 1640+10% foetal calf serum (FCS)+1 M 4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid (HEPES) buffer+200 mM L-glutamine) supplemented with IL-2 (50 U/ml; all from Sigma-Aldrich). At day 0, concanavalin A (2.5 mg/ml; Sigma-Aldrich) was added to the culture medium. The complete medium+IL-2 was renewed twice a week. Cells were collected at days 0, 3, 7, 14 and 21, washed with PBS and used directly for flow cytometry analysis.

T-cell population sorting

Mononuclear cells were separated from whole blood by centrifugation using a 1.077 density gradient media (Lymphoprep, Axis Shield PoC) and then stained using a direct immunofluorescence four-colour staining technique as described above. All experiments were performed on a BD FACSAria cell sorter using the BD FACSDiva software (Becton Dickinson).

The selective gating of different T-cell populations was based on the differential expression of the CD27, CD28 and CD45RO molecules. We used an anti-CD3 in conjunction with anti-CD28, anti-CD45RO and anti-CD27 mAbs that were conjugated with allophycocyanin (APC), phycoerythrin-Cy5 tandem (PC5), PE and fluorescein (FITC), respectively. This strategy allowed for a clear separation of the three previously mentioned blood T-cell populations: naive (CD27+ CD28+ CD45RO), activated/memory (CD27+ CD28+ CD45RO+) and effector (CD27 CD28) with more than 99.5% purity and without contamination between each population.

Cytokine-producing T-cell quantification

Previously sorted T-cell populations were cultured in a sterile environment in 15-ml tubes at a density of ~1 × 106 cells per ml for 5 h, at 37 °C under a 5% CO2 humidified atmosphere. Cells were cultured in complete medium (RPMI 1640+10% FCS+1 M HEPES buffer+200 mM L-glutamine) in the presence of brefeldin A (10 μg/ml), with (stimulated cultures) or without (un-stimulated controls) 12-O-tetradecanoylphorbol-13-acetate (PMA; 25 ng/ml) and ionomycin (1 μg/ml; all from Sigma-Aldrich).

The intracellular staining procedure was previously described in detail.40 In brief, the cells were harvested and stained for surface antigens and then fixed and permeabilised using the FIX & PERM Cell Fixation and Cell Permeabilization Kit (Life Technologies, Invitrogen, CA, USA). The cells were then stained with anti-IL-2, anti-IL-4, anti-IL-10, anti-interferon (IFN)-γ or anti-TNF-α mAbs conjugated with PE and analysed by flow cytometry.

Statistical analysis

All statistical analyses was performed using a one-way analysis of variance followed by Tukey’s honest significant difference test with a 95% confidence interval using the GraphPad prism 6.0 software (GraphPad Software Inc., La Jolla, CA, USA). Differences were considered statistically significant for P<0.05.

Ethical implications

All samples were collected after informed consent. This study was approved by the ethical review board of the Santo António Hospital, Porto, Portugal.


TCR-β gene diversity in circulating T cells from diabetic patients

We analysed the size of various PCR products that expanded parts of the TCR gene, encompassing the Vβ, Dβ and Jβ regions. The diversity in the PCR product sizes directly correlates with the V(D)J recombination diversity which, in turn, correlates with the T-cell clonal diversity, as previously described.37 The blood TCR-αβ+ T cells from non-diabetic individuals (controls; n=10; Figure 1a) presented a Gaussian distribution of the PCR product sizes, indicating that the TCR-αβ+ T cells from these individuals are clonally diverse. In contrast, the blood TCR-αβ+ T cells from diabetic individuals with acute DFU (n=10; Figure 1c) and from diabetic patients with chronic DFU (n=10; Figure 1d) clearly showed multiple clonal expansions represented by the presence of multiple peaks deviating from the Gaussian distribution.

Figure 1

Analysis of the Vβ-Jβ and Dβ-Jβ regions of the TCR-β gene locus in the PB of 10 healthy adult individuals (controls (a), 10 diabetic patients without DFU (b), 10 diabetic patients with acute DFU (<3 months of DFU (c)) and 10 diabetic patients with chronic DFU (>3 months of DFU (d)). The experiment was performed on isolated PB mononuclear cells from all individuals in each group. Because all results were similar, only one example is shown. The PCR product sizes were assessed by capillary electrophoresis on an ABI 310 electrophoresis system. A polyclonal T-cell distribution typically produces an approximately Gaussian profile of PCR product sizes. Peaks deviating from this Gaussian profile represent clonal expansion events and reflect a decreased diversity in the T-cell TCR-β gene rearrangements.

The diabetic patients without DFU (n=10; Figure 1b) showed no large clonal expansions but presented an intermediate phenotype between those of the controls and DFU patients. Indeed, small expansions were observed in 8 of the 10 diabetic patients analysed. However, all these patients displayed a significant decrease in the polyclonal background height (as shown in Figure 1b), which indirectly correlates with a reduction in the TCR-β repertoire diversity.37

In addition, we analysed the TCR-β repertoire diversity of diabetic patients who had previous episodes of DFU but had no active foot ulceration at the time of sample collection (n=3). In all cases, large TCR-β expansions were observed (data not shown).

Collectively, these results indicate that the TCR-β repertoire is less diverse in diabetic patients, especially in those with active foot ulcerations, and that this phenomenon persists after the DFU has healed.

Effect of the decreased circulating pool of naive T cells on inflammation

Because clonal expansion is a consequence of T-cell activation, we determined whether the various T-cell activation-related populations were contributing differently to the diminished TCR-β gene diversity observed in diabetic patients with DFU. We first quantified the relative representation of the most relevant T-cell activation-related populations in the PB of diabetic patients and controls, and then analysed the TCR-β gene rearrangement diversity in each population.

Three major T-cell populations were identified in the PB of controls and diabetic patients, and were defined on the basis of the relative expression of the co-stimulatory molecules CD27 and CD28, and the activation-associated molecule CD45RO: naive (CD27+ CD28+ CD45RO), activated/memory (CD27+ CD28+ CD45RO+) and effector (CD27 CD28). The gating strategy for these three populations is represented on Figure 2a.

Figure 2

Flow cytometry gating strategy (a) used for the identification of the most representative T-cell populations in the control and diabetic patients PB samples. Relative representation of the percentage of total CD4+ or CD8+ T cells in the naive (CD27+CD28+CD45RO) (b), activated/memory (CD27+CD28+CD45RO+) (c) and effector (CD27CD28) (d) stages in the PB of 22 healthy adult individuals (controls), 21 diabetic patients without DFU, 10 patients with acute DFU and 10 patients with chronic DFU. Data were compared between groups using a one-way analysis of variance followed by a Tukey’s honest significant difference post test with a 95% confidence interval using the GraphPad Prism 6.0 software. (*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001). DFU, diabetic foot ulceration; PB, peripheral blood.

We observed a significant decrease in the percentage of naive CD4+ T cells (Figure 2b) between controls (n=22) and all diabetic patient groups (n=41). This reduction of the percentage of naive CD4+ T cells in patients with chronic DFU (n=10) was not as severe as in the acute DFU patients (n=10), with a similar average reduction as in diabetic patients without DFU. In the case of CD8+ T cells, a similar trend was observed (Figure 2b), although the results were statistically significant in the acute DFU patients group only.

We also observed a significant increase in the percentage of activated/memory CD4+ T cells (Figure 2c) between controls and diabetic individuals. This tendency was not observed in the case of CD8+ T cells, in which no difference was observed.

In the case of effector T cells, we found a significant increase in the percentage of both CD4+ and CD8+ cells between diabetic patients with acute DFU and other diabetic patients or controls (Figure 2d). Although not statistically significant, we also observed a tendency towards an increase in the number of both CD4+ and CD8+ effector T cells in diabetic individuals without DFU compared with controls. Moreover, the diabetic patients without DFU showing the highest percentages of effector T cells (both CD4+ and CD8+) were also the ones showing the greatest reduction in their TCR-β repertoire diversity.

Interestingly, no difference was found in the effector T-cell percentages between diabetic patients with chronic DFU and both the diabetic patients without DFU and control groups (Figure 2d).

Effect of diabetes and foot ulceration on T cell chemokine receptor expression

Because both the chemokine environment11 and the immune cell migration capacity21 are altered in diabetic patients, we also investigated the expression of particular skin homing (CCR4), early inflammatory/Th1-specific (CCR5 and CXCR3) and late inflammatory (CXCR1) chemokine receptors (CHRs) on these T-cell populations.

In both the control and diabetic patient groups (n=6 in each group), the presence of CCR4 was almost restricted to the activated/memory T cells (P<0.01; Figure 3), with a higher expression level in the CD4+ subset. There was no significant difference in CCR4 expression between diabetic patients and controls.

Figure 3

Percentage of CCR4+, CCR5+, CXCR3+ and CXCR1+ cells in the PB naive, activated/memory and effector CD4+ and CD8+ T-cell populations from healthy adult individuals (controls) and diabetic patients without DFU or with acute and chronic DFU (n=6 in all four groups). The values represent the mean±s.d. Fresh blood cells were used to assess the CHR expression levels on each one of the T-cell populations by flow cytometry. No significant differences in CCR4 expression were observed between diabetic individuals and controls. In diabetic patients, the expression of CCR5 and CXCR3 was significantly reduced in the effector T cells (P<0.0001), and the expression of CXCR1 was only residual. CHR, chemokine receptor; DFU, diabetic foot ulceration; PB, peripheral blood.

CCR5 and CXCR3 were preferentially expressed in the activated/memory and effector T-cell populations, especially in the latter, in both diabetic patients and controls. Nonetheless, compared with controls, CCR5 and CXCR3 expression was significantly diminished in the effector T cells from the diabetic patients (P<0.0001) for both the CD4+ and CD8+ subsets. In addition, there was a significant decrease in the CCR5 expression level (P<0.001) in diabetic patients with chronic DFU when compared with the acute and non-DFU patients.

In contrast with the expression of the early inflammatory CHRs CCR5 and CXCR3 observed in all T-cell populations, the expression of the late inflammatory CHR CXCR1 was only detected in the CD4+ or CD8+ effector T cells, with a consistently higher expression level in the CD8+ cells. Interestingly, almost no CXCR1 expression was observed in the effector T cells from diabetic individuals with or without foot ulceration (P<0.0001).

Furthermore, we analysed the effect of excessive stimulation on the CHR expression levels of T cells in vitro to mimic the excessive inflammatory conditions observed in diabetic patients. Therefore, we stimulated mononuclear cells from non-diabetic individuals (controls; n=6) with concanavalin A in the presence of IL-2 and cultured these cells for 3 weeks. The expression of the previously analysed CHRs (CCR4, CCR5, CXCR1 and CXCR3) was monitored on days 0, 3, 7, 14 and 21 (Figure 4).

Figure 4

Percentage of CCR4+, CCR5+, CXCR1+ and CXCR3+ T cells after in vitro stimulation. The values represent the mean±s.d. Mononuclear cells were isolated from the blood of six healthy adult individuals and were cultured during 3 weeks. At day 0, the cells were stimulated with concanavalin-A and IL-2. CHR expression was assessed on T cells by flow cytometry on days 0, 3, 7, 14 and 21. In all samples, the percentage of CCR4+ and CXCR3+ T cells increased, whereas the percentage of CCR5+ and CXCR1+ T cells decreased. Only the decrease in CCR5 expression was statistically significant. CHR, chemokine receptor; IL, interleukin.

Under these conditions, the percentage of T cells expressing CCR4 and CXCR3 increased consistently during the 3 weeks of culture, although this increase was not statistically significant. Conversely, the expression of CXCR1 and CCR5 decreased during the 21 days of culture, a change that was significant only for CCR5 expression (P<0.01).

The T-cell activation profile limits TCR clonal diversity

We showed that diabetic patients with acute or chronic DFU display a skewed TCR-Vβ repertoire (Figure 1). In order to evaluate whether this is a cell-intrinsic mechanism associated with the differentiation of T cells into effector cells, we used the same methodology as before to analyse the TCR-β gene rearrangement diversity in sorted naive, activated/memory and effector T-cell populations from the PB of six healthy adult individuals. We also analysed the TCR-Vβ family representation at the protein level by flow cytometry.

We observed that both the naive (Figure 5a) and activated/memory T-cell pools (Figure 5b) expressed a wide range of TCR-Vβ families, which correlates with a high level of diversity in the TCR-Vβ family repertoire. In contrast, the effector population displayed a clear tendency towards the overexpression of certain TCR-Vβ families (different within individuals), with the concomitant absence of others (Figure 5c).

Figure 5

TCR-Vβ family repertoire of the naive (CD27+CD28+CD45RO) (a); activated/memory (CD27+CD28+CD45RO+) (b); and effector (CD27CD28) (c) PB T-cell populations from six healthy adult individuals (blood donors) assessed by flow cytometry. Analysis of the Vβ-Jβ and Dβ-Jβ regions of the TCR-β gene locus on the naive (d), activated/memory (e) and effector (f) populations sorted from the PB of six healthy adult individuals reflecting TCR-β gene rearrangement diversity. All samples produced a similar result, and therefore only one example is shown. The PCR product sizes were assessed by single-capillary electrophoresis on an ABI 310 apparatus. A polyclonal T-cell distribution produces an approximately Gaussian profile of the PCR product sizes. The peaks deviating from this typical Gaussian profile represent clonal expansions and reflect a decreased diversity in the T-cell TCR-Vβ gene repertoire. The naive and activated/memory cells displayed a polyclonal TCR-Vβ gene repertoire, whereas the effector T cells presented multiple clonal expansions and concomitant gaps in the TCR-Vβ gene repertoire. PB, peripheral blood; PCR, polymerase chain reaction; TCR, T-cell receptor.

At the gene rearrangement level, we observed that the PCR product sizes from both the naive (Figure 5d) and activated/memory (Figure 5e) subpopulations followed a Gaussian distribution, reflecting a polyclonal TCR-β gene rearrangement. Again, the effector population (Figure 5f) significantly deviated from this Gaussian distribution by displaying a significant reduction in the TCR-β gene diversity in both the Vβ-Jβ and Dβ-Jβ regions.

Because, in diabetic individuals, the T cells have an important role in the maintenance of the systemic pro-inflammatory environment41 that is partially responsible for their impaired wound healing,11 we analysed the production of key inflammatory cytokines (IL-2, IFN-γ and TNF-α) by each of the sorted T-cell populations.

The naive and activated/memory T cells from the CD4+ (Figure 6a) or CD8+ (Figure 6b) subsets were both able to secrete IL-2 upon stimulation with PMA and ionomycin. However, only a minimal IL-2 production level was observed in effector T cells. The percentage of IL-2-secreting cells was higher in the activated/memory T cell than in the naive T-cell pools (P<0.01 for CD8; P<0.05 for CD4).

Figure 6

Percentage of IL-2, IFN-γ and TNF-α expressing cells among sorted naive, activated/memory and effector populations in both CD4+ (a) and CD8+ (b) T cells assessed by flow cytometry. The sorted naive, activated/memory and effector T-cell populations from six healthy adult individuals were stimulated with PMA+ionomycin for 5 h before analysis. The synthesis of IFN-γ and TNF-α was not observed in naive T cells, but increased during T-cell stimulation. IL-2 synthesis was observed in all T-cell populations, but was minimal in effector T cells. IFN, interferon; IL, interleukin; PMA, 12-O-tetradecanoylphorbol-13-acetate; TNF, tumour necrosis factor.

In contrast, both IFN-γ and TNF-α expression levels were only detected on the activated/memory and effector T cells, with a significantly higher expression level of IFN-γ in the effector T-cell population compared with the activated/memory population (P<0.01; Figure 6).


The early studies involving murine models42, 43 unravelled a dual role of T cells in both the inflammatory and proliferation phases of wound healing. Although wound healing impairment has usually been associated with the effect of cytotoxic CD8+ T cells, the subset of T cells responsible for wound healing enhancement remained to be identified.44 Using a different approach to this problem, we analysed how diabetes, in general, and DFU, in particular, affect TCR diversity. We also analysed the relative distribution of the most representative T-cell populations.

We demonstrated, for the first time, that DFU patients show a significantly reduced blood T-cell TCR-β repertoire diversity that directly correlates with both a decrease in the naive T cell and a concomitant increase in the effector T-cell pools. Our data clearly show that effector T cells display a significantly reduced TCR-β repertoire diversity. Because T cells tend to accumulate in diabetic patients (especially in those with acute DFU), the overall TCR-β repertoire diversity is reduced in these patients. The BMI of the diabetic patients and controls was similar (data not shown), thereby excluding obesity as the cause for the restricted TCR-β repertoire diversity.

Based on our data, we cannot infer that the TCR-β diversity observed in these patients was decreased to the point where the T-cell pool would fail to recognize antigens, which could, as previously observed in aged individuals, impair the immune response to subsequent infections by different pathogens.27, 33, 34 Nevertheless, our data indicate that a reduced TCR-β diversity is present in most diabetic patients, is more severe in DFU patients and persists after DFU healing. Further experiments are needed to construct evidence-based mathematical models that could predict the effects of TCR-β restriction on the immune function. Longitudinal studies are underway to determine whether, as expected, diabetic patients with a significantly reduced TCR-β repertoire diversity are more prone to DFU.

Surprisingly, we also found that diabetic patients with chronic foot ulceration usually have a similar naive and effector T-cell pool size to those of diabetic patients without foot ulcers. However, a considerably larger cohort will be needed to understand the reasons behind this observation because the chronic DFU patient cohort was very heterogeneous compared with the acute DFU cohort. Indeed, the chronic DFU cohort comprised patients with wounds at different stages, patients who were already in remission or patients who were given different treatments.

Our results also suggest that diabetes alone does not cause a severe reduction in the TCR-β diversity, presumably because the systemic pro-inflammatory environment is not sufficient to fully drive T-cell activation and promote clonal expansion. Nevertheless, the pro-inflammatory environment observed in diabetic patients seems to enhance clonal expansion, leading to a significant decrease in the naive T-cell subset with an accumulation of T cells at various stages of activation. In fact, some diabetic individuals who do not have active or previous foot ulcerations, already have a significant increase in the percentage and number (data not shown) of effector T cells and display various large clonal expansions. This suggests that the accumulation of effector T cells and the concomitant reduction in the overall TCR-Vβ diversity represent a continuous phenomenon driven by diabetes.

Alternatively, our results partially explain the reason behind the self-sustaining systemic inflammatory environment in diabetics. We have shown that the effector T cells, which tend to accumulate in diabetic patients, are the major producers of the IFN-γ and TNF-α inflammatory cytokines that, in turn, enhance naive T-cell activation and differentiation.45, 46 Importantly, activated/memory T cells also contribute to the observed systemic inflammatory environment to a smaller extent by producing lower levels of IFN-γ and TNF-α. TNF-α alone has been shown to have various harmful effects on diabetic wound healing, such as the inhibition of fibroblasts and keratinocytes proliferation and migration,47, 48 the induction of apoptosis in endothelial cells and pericytes, or the increase in the expression of the FOXO1 transcription factors.49, 50 In fact, the inhibition of TNF-α in vivo significantly improves wound closure in animal models.51 Our group has already demonstrated that neurotensin, either in vitro52, 53 or when used in wound dressings,54 inhibits TNF-α expression, reduces inflammation and promotes wound healing in diabetic mice. Our results also partly explain why T-cell migration is altered independently of the presence of foot ulcerations in diabetic patients.

Although we did not observe any difference in the expression of the skin homing receptor CCR4 between the T cells from diabetic individuals and controls, the expression of the inflammatory CHRs CCR5, CXCR1 and CXCR3 was significantly altered. In non-diabetic individuals, the expression of the early inflammatory/Th1-related CHRs CCR5 and CXCR3 is observed in both activated/memory and effector T cells, whereas the expression of CXCR1 is observed only in the effector T cells. In the T cells from diabetic patients, a significant reduction in the expression of these CHRs was observed, including in patients without DFU.

The in vitro stimulation assays mimicking the pro-inflammatory environment observed in diabetes revealed a reduction in the CCR5 and CXCR1 expression levels in T cells. In contrast, a clear increase in CXCR3 expression was observed after T-cell stimulation. The internalisation of CXCR3 by IFN-γ-activated venous endothelial cells (as observed in diabetic patients) has already been described.55 Because our cultures only contained blood mononuclear cells, this effect could not be observed and might explain the differences observed between the in vitro and in vivo CXCR3 expression changes.

We do not yet understand how and why the expression of these CHRs is reduced, but, collectively with previous studies, our results lead us to speculate that overstimulation could promote their internalisation.56, 57 Nevertheless, the profound reduction in the expression of these CHRs on the T cells from diabetic patients is expected to adversely impact T-cell migration to inflamed tissues such as diabetic foot ulcers.

In conclusion, our results strongly emphasize the dysfunctional immune response observed in diabetic patients. For the first time, we have analysed the effect of diabetes on the TCR repertoire diversity in circulating TCR-αβ+ T cells and on the distribution of T-cell populations. Our results showed that diabetes has a profound impact on the circulating T-cell pool, increasing the naive/effector T-cell ratio and reducing TCR-β repertoire diversity.

Conversely, we have shown that the concomitant and continuous accumulation of both CD4+ and CD8+ effector T cells may be responsible for the abnormally high IFN-γ and TNF-α levels observed in diabetic patients, and may lead to a reduction of the inflammatory CHR expression, which could possibly affect T-cell migration into inflamed tissues.

Because the accumulation of effector T cells appears central to the DFU pathology, immunotherapeutic strategies should be devised in order to diminish T-cell activation and tissue accumulation. Moreover, the number of effector T cells and the overall TCR-β repertoire diversity may hold a prognostic value in the treatment of DFU and possibly other diabetes-associated complications.


  1. 1

    Moura LI, Dias AM, Carvalho E, de Sousa HC . Recent advances on the development of wound dressings for diabetic foot ulcer treatment—a review. Acta Biomater 2013; 9: 7093–7114.

  2. 2

    Moura J, Borsheim E, Carvalho E . The role of microRNAs in diabetic complications-special emphasis on wound healing. Genes 2014; 5: 926–956.

  3. 3

    Kavitha KV, Tiwari S, Purandare VB, Khedkar S, Bhosale SS, Unnikrishnan AG . Choice of wound care in diabetic foot ulcer: a practical approach. World J Diabetes 2014; 5: 546–556.

  4. 4

    Berlanga-Acosta J . Diabetic lower extremity wounds: the rationale for growth factors-based infiltration treatment. Int Wound J 2011; 8: 612–620.

  5. 5

    Drela E, Stankowska K, Kulwas A, Rosc D . Endothelial progenitor cells in diabetic foot syndrome. Adv Clin Exp Med 2012; 21: 249–254.

  6. 6

    Kirana S, Stratmann B, Prante C, Prohaska W, Koerperich H, Lammers D et al. Autologous stem cell therapy in the treatment of limb ischaemia induced chronic tissue ulcers of diabetic foot patients. Int J Clin Pract 2012; 66: 384–393.

  7. 7

    Al-Wahbi AM . Impact of a diabetic foot care education program on lower limb amputation rate. Vasc Health Risk Manag 2010; 6: 923–934.

  8. 8

    Doupis J, Veves A . Classification, diagnosis, and treatment of diabetic foot ulcers. Wounds 2008; 20: 117–126.

  9. 9

    Kishore S, Upadhyay AD, Jyotsna VP . Categories of foot at risk in patients of diabetes at a tertiary care center: insights into need for foot care. Indian J Endocrinol Metab 2015; 19: 405–410.

  10. 10

    Dinh T, Tecilazich F, Kafanas A, Doupis J, Gnardellis C, Leal E et al. Mechanisms involved in the development and healing of diabetic foot ulceration. Diabetes 2012; 61: 2937–2947.

  11. 11

    Acosta JB, del Barco DG, Vera DC, Savigne W, Lopez-Saura P, Guillen Nieto G et al. The pro-inflammatory environment in recalcitrant diabetic foot wounds. Int Wound J 2008; 5: 530–539.

  12. 12

    Tuttolomondo A, Maida C, Pinto A . Diabetic foot syndrome as a possible cardiovascular marker in diabetic patients. J Diabetes Res 2015; 2015: 268390.

  13. 13

    Tellechea A, Kafanas A, Leal EC, Tecilazich F, Kuchibhotla S, Auster ME et al. Increased skin inflammation and blood vessel density in human and experimental diabetes. Int J Low Extrem Wounds 2013; 12: 4–11.

  14. 14

    Tuttolomondo A, Maida C, Pinto A . Diabetic foot syndrome: Immune-inflammatory features as possible cardiovascular markers in diabetes. World J Orthop 2015; 6: 62–76.

  15. 15

    Pinto A, Tuttolomondo A, Di Raimondo D, Fernandez P, La Placa S, Di Gati M et al. Cardiovascular risk profile and morbidity in subjects affected by type 2 diabetes mellitus with and without diabetic foot. Metabolism 2008; 57: 676–682.

  16. 16

    Pinto A, Tuttolomondo A, Di Raimondo D, La Placa S, Di Sciacca R, Fernandez P et al. Ischemic stroke in patients with diabetic foot. Int Angiol 2007; 26: 266–269.

  17. 17

    Tuttolomondo A, La Placa S, Di Raimondo D, Bellia C, Caruso A, Lo Sasso B et al. Adiponectin, resistin and IL-6 plasma levels in subjects with diabetic foot and possible correlations with clinical variables and cardiovascular co-morbidity. Cardiovasc Diabetol 2010; 9: 50.

  18. 18

    Otranto M, Nascimento AP, Monte-Alto-Costa A . Insulin resistance impairs cutaneous wound healing in mice. Wound Repair Regen 2013; 21: 464–472.

  19. 19

    Dickinson S, Hancock DP, Petocz P, Ceriello A, Brand-Miller J . High-glycemic index carbohydrate increases nuclear factor-kappaB activation in mononuclear cells of young, lean healthy subjects. Am J Clin Nutr 2008; 87: 1188–1193.

  20. 20

    Stegenga ME, van der Crabben SN, Dessing MC, Pater JM, van den Pangaart PS, de Vos AF et al. Effect of acute hyperglycaemia and/or hyperinsulinaemia on proinflammatory gene expression, cytokine production and neutrophil function in humans. Diabet Med 2008; 25: 157–164.

  21. 21

    Bogdanski P, Pupek-Musialik D, Dytfeld J, Jagodzinski PP, Jablecka A, Kujawa A et al. Influence of insulin therapy on expression of chemokine receptor CCR5 and selected inflammatory markers in patients with type 2 diabetes mellitus. Int J Clin Pharmacol Ther 2007; 45: 563–567.

  22. 22

    Kaser S, Kaser A, Sandhofer A, Ebenbichler CF, Tilg H, Patsch JR . Resistin messenger-RNA expression is increased by proinflammatory cytokines in vitro. BiochemBiophys Res Commun 2003; 309: 286–290.

  23. 23

    Gardner SE, Hillis SL, Heilmann K, Segre JA, Grice EA . The neuropathic diabetic foot ulcer microbiome is associated with clinical factors. Diabetes 2013; 62: 923–930.

  24. 24

    Mi Q, Riviere B, Clermont G, Steed DL, Vodovotz Y . Agent-based model of inflammation and wound healing: insights into diabetic foot ulcer pathology and the role of transforming growth factor-beta1. Wound Repair Regen 2007; 15: 671–682.

  25. 25

    Davis PA, Corless DJ, Aspinall R, Wastell C . Effect of CD4(+) and CD8(+) cell depletion on wound healing. Br J Surg 2001; 88: 298–304.

  26. 26

    Loots MA, Lamme EN, Zeegelaar J, Mekkes JR, Bos JD, Middelkoop E . Differences in cellular infiltrate and extracellular matrix of chronic diabetic and venous ulcers versus acute wounds. J Investig Dermatol 1998; 111: 850–857.

  27. 27

    Brunner S, Herndler-Brandstetter D, Weinberger B, Grubeck-Loebenstein B . Persistent viral infections and immune aging. Ageing Res Rev 2011; 10: 362–369.

  28. 28

    Viallard JF, Ruiz C, Guillet M, Pellegrin JL, Moreau JF . Perturbations of the CD8(+) T-cell repertoire in CVID patients with complications. Results Immunol 2013; 3: 122–128.

  29. 29

    Kharbanda M, McCloskey TW, Pahwa R, Sun M, Pahwa S . Alterations in T-cell receptor Vbeta repertoire of CD4 and CD8 T lymphocytes in human immunodeficiency virus-infected children. Clin Diagn Lab Immunol 2003; 10: 53–58.

  30. 30

    Luo W, Su J, Zhang XB, Yang Z, Zhou MQ, Jiang ZM et al. Limited T cell receptor repertoire diversity in tuberculosis patients correlates with clinical severity. PloS One 2012; 7: e48117.

  31. 31

    Xiong Y, Tan Y, Song YG . Analysis of T cell receptor Vbeta diversity in peripheral CD4+ and CD8+ T lymphocytes obtained from patients with chronic severe hepatitis B. Hepat Mon 2014; 14: e15900.

  32. 32

    Stockinger B, Barthlott T, Kassiotis G . The concept of space and competition in immune regulation. Immunology 2004; 111: 241–247.

  33. 33

    Yager EJ, Ahmed M, Lanzer K, Randall TD, Woodland DL, Blackman MA . Age-associated decline in T cell repertoire diversity leads to holes in the repertoire and impaired immunity to influenza virus. J Exp Med 2008; 205: 711–723.

  34. 34

    Blackman MA, Woodland DL . The narrowing of the CD8 T cell repertoire in old age. Curr Opin Immunol 2011; 23: 537–542.

  35. 35

    Armstrong KM, Piepenbrink KH, Baker BM . Conformational changes and flexibility in T-cell receptor recognition of peptide-MHC complexes. Biochem J 2008; 415: 183–196.

  36. 36

    Werdin F, Tennenhaus M, Schaller HE, Rennekampff HO . Evidence-based management strategies for treatment of chronic wounds. Eplasty 2009; 9: e19.

  37. 37

    van Dongen JJ, Langerak AW, Bruggemann M, Evans PA, Hummel M, Lavender FL et al. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 concerted action BMH4-CT98-3936. Leukemia 2003; 17: 2257–2317.

  38. 38

    Renzi P, Ginns LC . Analysis of T cell subsets in normal adults. Comparison of whole blood lysis technique to Ficoll-Hypaque separation by flow cytometry. J Immunol Methods 1987; 98: 53–56.

  39. 39

    Appay V, van Lier RA, Sallusto F, Roederer M . Phenotype and function of human T lymphocyte subsets: consensus and issues. Cytometry Part A 2008; 73: 975–983.

  40. 40

    Campana D, Thompson JS, Amlot P, Brown S, Janossy G . The cytoplasmic expression of CD3 antigens in normal and malignant cells of the T lymphoid lineage. J Immunol 1987; 138: 648–655.

  41. 41

    Mirza RE, Koh TJ . Contributions of cell subsets to cytokine production during normal and impaired wound healing. Cytokine 2015; 71: 409–412.

  42. 42

    Barbul A, Regan MC . The regulatory role of T lymphocytes in wound healing. J Trauma 1990; 30: S97–100.

  43. 43

    Efron JE, Frankel HL, Lazarou SA, Wasserkrug HL, Barbul A . Wound healing and T-lymphocytes. J Surg Res 1990; 48: 460–463.

  44. 44

    Barbul A, Breslin RJ, Woodyard JP, Wasserkrug HL, Efron G . The effect of in vivo T helper and T suppressor lymphocyte depletion on wound healing. Ann Surg 1989; 209: 479–483.

  45. 45

    Aspalter RM, Eibl MM, Wolf HM . Regulation of TCR-mediated T cell activation by TNF-RII. J Leukoc Biol 2003; 74: 572–582.

  46. 46

    Watts TH . Staying alive: T cell costimulation, CD28, and Bcl-xL. J Immunol 2010; 185: 3785–3787.

  47. 47

    Hasnan J, Yusof MI, Damitri TD, Faridah AR, Adenan AS, Norbaini TH . Relationship between apoptotic markers (Bax and Bcl-2) and biochemical markers in type 2 diabetes mellitus. Singapore Med J 2010; 51: 50–55.

  48. 48

    Ruckert R, Lindner G, Bulfone-Paus S, Paus R . High-dose proinflammatory cytokines induce apoptosis of hair bulb keratinocytes in vivo. Br J Dermatol 2000; 143: 1036–1039.

  49. 49

    Ponugoti B, Dong G, Graves DT . Role of forkhead transcription factors in diabetes-induced oxidative stress. Exp Diabetes Res 2012; 2012: 939751.

  50. 50

    Potente M, Urbich C, Sasaki K, Hofmann WK, Heeschen C, Aicher A et al. Involvement of Foxo transcription factors in angiogenesis and postnatal neovascularization. J Clin Investig 2005; 115: 2382–2392.

  51. 51

    Siqueira MF, Li J, Chehab L, Desta T, Chino T, Krothpali N et al. Impaired wound healing in mouse models of diabetes is mediated by TNF-alpha dysregulation and associated with enhanced activation of forkhead box O1 (FOXO1). Diabetologia 2010; 53: 378–388.

  52. 52

    Moura LI, Silva L, Leal EC, Tellechea A, Cruz MT, Carvalho E . Neurotensin modulates the migratory and inflammatory response of macrophages under hyperglycemic conditions. BioMed Res Int 2013; 2013: 941764.

  53. 53

    da Silva L, Neves BM, Moura L, Cruz MT, Carvalho E . Neurotensin downregulates the pro-inflammatory properties of skin dendritic cells and increases epidermal growth factor expression. Biochim Biophys Acta 2011; 1813: 1863–1871.

  54. 54

    Moura LI, Dias AM, Leal EC, Carvalho L, de Sousa HC, Carvalho E . Chitosan-based dressings loaded with neurotensin—an efficient strategy to improve early diabetic wound healing. Acta Biomater 2014; 10: 843–857.

  55. 55

    Sauty A, Colvin RA, Wagner L, Rochat S, Spertini F, Luster AD . CXCR3 internalization following T cell-endothelial cell contact: preferential role of IFN-inducible T cell alpha chemoattractant (CXCL11). J Immunol 2001; 167: 7084–7093.

  56. 56

    Rose JJ, Foley JF, Murphy PM, Venkatesan S . On the mechanism and significance of ligand-induced internalization of human neutrophil chemokine receptors CXCR1 and CXCR2. J Biol Chem 2004; 279: 24372–24386.

  57. 57

    Maritzen T, Schachtner H, Legler DF . On the move: endocytic trafficking in cell migration. Cell Mol Life Sci 2015; 72: 2119–2134.

Download references


This work was financed by FEDER funds by the operational program Factors of Competitivity – COMPETE, by the Portuguese Foundation for Science and Technology (FCT) - EXCL/DTP-PIC/0069/2012, PEst-C/SAU/LA0001/2013 and UID/NEU/04539/2013, the EFSD European Research Programme in Microvascular Complications of Diabetes supported by Novartis, and Forum Hematológico do Norte, Portugal. Eugénia Carvalho is partly funded by the Arkansas Biosciences Institute, the major research component of the Arkansas Tobacco Settlement Proceeds Act of 2000, NIH P30AG028718, and NIH RO1 AG033761.

Author information

Correspondence to João Moura.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark


  • diabetes
  • T-cells
  • T-cell repertoire diversity
  • wound healing

Further reading