**5. Familial cardiovascular abnormalities**

Diabetes and hypertension are strongly associated although the role of glycaemia in promoting hypertension is a matter of debate (Invitti, 2003). HbA1c variability predicts not only incident microalbuminuria and progression of established renal disease but also cardiovascular disease events in patients with type 1 diabetes (Wadén et al., 2009). Moreover, HbA1c concentration predicts cardiovascular disease and all-cause mortality in adults without diabetes (Khaw et al., 2004). In healthy non-diabetic and non-hypertensive men, fasting plasma glucose is independently associated with blood pressure at rest and during exercise and development of elevated blood pressure after 7-years follow-up (Biornholt et al., 2003). Usually, in type 1 diabetes families, parental hypertension has been associated with diabetic nephropathy in adult and young offspring (Viberti et al., 1987; Marcovecchio et al., 2010), but the familial/hereditary factors that have an impact on diabetic nephropathy have not been so far identified. In a large homogeneous population from the Finnish Diabetic Nephropathy study, a cluster of parental hypertension, cardiovascular disease, cardiovascular mortality, and type 1 diabetes was associated with diabetic nephropathy in offspring with type 1 diabetes. It seemed that the more the traits clustered in family, the higher the risk for diabetes nephropathy (Thorn et al., 2007).

In this regard it is noteworthy that enhanced sodium/lithium countertransport and sodium/hydrogen exchange had been suggested to predict diabetic nephropathy (Walker et al., 1990; Ng et al, 1990). However, we found evidence contradicting this favourite hypothesis. Indeed, our data demonstrated convincingly that sodium/hydrogen exchange activity was significantly higher in type 1 diabetes with no difference among the two groups

The Enlarging List of Phenotypic Characteristics That

groups (Matteucci et al., 2010).

(Matteucci et al., 2010).

electron transport systems in vascular pathobiology.

Might Allow the Clinical Identification of Families at Risk for Type 1 Diabetes 313

findings supported, in a clinical setting, the proposed role of transplasma membrane

Fig. 2. Homeostasis model assessment of insulin sensitivity (HOMA-IS) in the same study

Fig. 3. Systolic blood pressure amplitude (SBP-Amplitude) in the same study groups

**6. Biochemical phenotype and redox balance in type 1 diabetes relatives** 

Our studies over the years have linked family history of type 1 diabetes (first-degree kinship) with multiple biochemical abnormalities. Since 2000 we documented metabolic perturbations in nondiabetic relatives: parents differed from age-matched control subjects in the higher plasma concentrations of glucose and Lipoprotein (a); their fibrinogen was borderline but did not reach any statistical significance; in turn, siblings of type 1 diabetes

of diabetic patients with and without nephropathy. Moreover, enhanced sodium/hydrogen exchange activity was also a common feature of nondiabetic first-degree relatives of type 1 diabetic patients with no difference among the corresponding groups of relatives. The association between antiport activities of diabetic probands and their relatives suggested that the altered activity of the transporter was primarily determined by familial factors whose nature remained to be clarified (Matteucci & Giampietro, 2000b).

Generally, the observation of raised arterial blood pressure in relatives of type 1 diabetes patients was based on history, a single measurement of arterial blood pressure, or a 24-h ambulatory record; we were first to evaluate the response to ergometer exercise (Matteucci et al., 2006). Blood pressure response to exercise had been evaluated as a predictor of future hypertension and cardiovascular disease (Sharabi et al., 2001). Moreover, the heritability for resting blood pressure and blood pressure response to exercise was under investigation (An et al., 2000). We identified an abnormal blood pressure response to exercise testing not only in type 1 diabetic probands but also in asymptomatic normotensive non-diabetic relatives of type 1 diabetics, in which it was associated with indices of metabolic syndrome and oxidative damage. Furthermore, in healthy normotensive non-diabetic control subjects without family history of type 1 diabetes, strong associations were found 1) between resting systolic blood pressure and fasting plasma glucose as well as fasting plasma insulin levels, and 2) between systolic blood pressure response to exercise and HbA1c levels (Matteucci et al., 2006).

In a recent study, we performed 24-hour ambulatory blood pressure monitoring in type 1 diabetes families with the primary aim of investigating the circadian variability of blood pressure and the ambulatory arterial stiffness index in healthy siblings of type 1 diabetes patients vs healthy control subjects who had no first-degree relative with type 1 diabetes (Matteucci et al., 2010). Secondary aims of the study were to explore the influence of both cardiovascular autonomic function and erythrocyte electron transfer activity as oxidative marker on the ambulatory blood pressure profile. Indeed, human erythrocytes possess a transplasma ferricyanide reductase activity (measured as the erythrocyte velocity of ferricyanide reduction) that transfers reducing equivalents from intracellular reductants to extracellular oxidants (Matteucci & Giampietro, 2000c) and belongs to the ubiquitous transplasma membrane electron transport systems. Transplasma membrane electron transport activities have been related to the regulation of vital cellular processes and to the pathogenesis of various human disorders (Lane & Lawen, 2009) and exist also in endothelial cells where they have been suggested to regulate redox status and possibly atherogenesis through regulation of haeme oxygenase-1 expression (Lee et al., 2009).

We found that systolic blood pressure midline-estimating statistic of rhythm and pulse pressure were higher in type 1 diabetes patients and correlated positively with diabetes duration and the rate of oxidant-induced erythrocyte electron transfer to extracellular ferricyanide. Autonomic dysfunction was associated with diastolic blood pressure ecphasia and increased ambulatory arterial stiffness index. Siblings had higher BMI (Figure 1), lower insulin sensitivity (Figure 2), larger systolic blood pressure amplitude (Figure 3), and higher ambulatory arterial stiffness index than controls. Daytime systolic blood pressure was positively, independently associated with BMI and erythrocyte electron transfer to extracellular ferricyanide. Among non-diabetic people, there was a significant correlation between ambulatory arterial stiffness index and fasting plasma glucose. We concluded that siblings of type 1 diabetes patients exhibited a cluster of sub-clinical metabolic abnormalities associated with consensual perturbations in blood pressure variability. Moreover, our

of diabetic patients with and without nephropathy. Moreover, enhanced sodium/hydrogen exchange activity was also a common feature of nondiabetic first-degree relatives of type 1 diabetic patients with no difference among the corresponding groups of relatives. The association between antiport activities of diabetic probands and their relatives suggested that the altered activity of the transporter was primarily determined by familial factors

Generally, the observation of raised arterial blood pressure in relatives of type 1 diabetes patients was based on history, a single measurement of arterial blood pressure, or a 24-h ambulatory record; we were first to evaluate the response to ergometer exercise (Matteucci et al., 2006). Blood pressure response to exercise had been evaluated as a predictor of future hypertension and cardiovascular disease (Sharabi et al., 2001). Moreover, the heritability for resting blood pressure and blood pressure response to exercise was under investigation (An et al., 2000). We identified an abnormal blood pressure response to exercise testing not only in type 1 diabetic probands but also in asymptomatic normotensive non-diabetic relatives of type 1 diabetics, in which it was associated with indices of metabolic syndrome and oxidative damage. Furthermore, in healthy normotensive non-diabetic control subjects without family history of type 1 diabetes, strong associations were found 1) between resting systolic blood pressure and fasting plasma glucose as well as fasting plasma insulin levels, and 2) between systolic

whose nature remained to be clarified (Matteucci & Giampietro, 2000b).

blood pressure response to exercise and HbA1c levels (Matteucci et al., 2006).

through regulation of haeme oxygenase-1 expression (Lee et al., 2009).

In a recent study, we performed 24-hour ambulatory blood pressure monitoring in type 1 diabetes families with the primary aim of investigating the circadian variability of blood pressure and the ambulatory arterial stiffness index in healthy siblings of type 1 diabetes patients vs healthy control subjects who had no first-degree relative with type 1 diabetes (Matteucci et al., 2010). Secondary aims of the study were to explore the influence of both cardiovascular autonomic function and erythrocyte electron transfer activity as oxidative marker on the ambulatory blood pressure profile. Indeed, human erythrocytes possess a transplasma ferricyanide reductase activity (measured as the erythrocyte velocity of ferricyanide reduction) that transfers reducing equivalents from intracellular reductants to extracellular oxidants (Matteucci & Giampietro, 2000c) and belongs to the ubiquitous transplasma membrane electron transport systems. Transplasma membrane electron transport activities have been related to the regulation of vital cellular processes and to the pathogenesis of various human disorders (Lane & Lawen, 2009) and exist also in endothelial cells where they have been suggested to regulate redox status and possibly atherogenesis

We found that systolic blood pressure midline-estimating statistic of rhythm and pulse pressure were higher in type 1 diabetes patients and correlated positively with diabetes duration and the rate of oxidant-induced erythrocyte electron transfer to extracellular ferricyanide. Autonomic dysfunction was associated with diastolic blood pressure ecphasia and increased ambulatory arterial stiffness index. Siblings had higher BMI (Figure 1), lower insulin sensitivity (Figure 2), larger systolic blood pressure amplitude (Figure 3), and higher ambulatory arterial stiffness index than controls. Daytime systolic blood pressure was positively, independently associated with BMI and erythrocyte electron transfer to extracellular ferricyanide. Among non-diabetic people, there was a significant correlation between ambulatory arterial stiffness index and fasting plasma glucose. We concluded that siblings of type 1 diabetes patients exhibited a cluster of sub-clinical metabolic abnormalities associated with consensual perturbations in blood pressure variability. Moreover, our findings supported, in a clinical setting, the proposed role of transplasma membrane electron transport systems in vascular pathobiology.

Fig. 2. Homeostasis model assessment of insulin sensitivity (HOMA-IS) in the same study groups (Matteucci et al., 2010).

Fig. 3. Systolic blood pressure amplitude (SBP-Amplitude) in the same study groups (Matteucci et al., 2010).

#### **6. Biochemical phenotype and redox balance in type 1 diabetes relatives**

Our studies over the years have linked family history of type 1 diabetes (first-degree kinship) with multiple biochemical abnormalities. Since 2000 we documented metabolic perturbations in nondiabetic relatives: parents differed from age-matched control subjects in the higher plasma concentrations of glucose and Lipoprotein (a); their fibrinogen was borderline but did not reach any statistical significance; in turn, siblings of type 1 diabetes

The Enlarging List of Phenotypic Characteristics That

to be prevailing.

Taken into account that:

et al., 2007),

al., 2004),

insulin-stimulated muscle glucose disposal.

hyperpolarisation of the mitochondrial membrane.

Might Allow the Clinical Identification of Families at Risk for Type 1 Diabetes 315

(Matteucci et al., 2001). We were unable to find out any abnormalities in circulating metal ions (such as iron, transferrin, ferritin, copper, and ceruloplasmin) or extracellular antioxidant defences (such as serum uric acid, albumin, bilirubin,) that could favour oxidative stress in non-diabetic relatives of type 1 patients. On the contrary, we confirmed our previous finding of a generalised increase in sodium/hydrogen exchange activity. The rate of amiloride-sensitive hydrogen efflux from erythrocytes was significantly associated with both erythrocyte glutathione content and some markers of radical-induced damage such as plasma advanced oxidation protein products and malondialdehyde, erythrocyte osmotic fragility, and erythrocyte malondialdehyde accumulation under oxidative stress. Hence, this additional study provided the first *in vivo* demonstration of a significant association between oxidative stress and sodium/hydrogen exchange upregulation. The familiarly overactive sodium/hydrogen exchange itself could be viewed as further evidence pointing to the presence in these families of a redox disequilibrium where oxidation seems

 mitochondria are the cellular site of oxidation-reduction reactions and energy transfer processes; mitochondrial dysfunction is believed to play a role in the development of diabetes and its complications because of the active generation of free radicals (Maiese

 a reactive oxygen species-mediated long-term 'memory' of hyperglycaemic stress has been reported in the mitochondria of endothelial cells (Ihnat et al., 2007), but impairment of mitochondrial function has been also observed in subjects with family history of type 2 diabetes before the onset of impaired glucose tolerance (Petersen et

in the last step of our research we measured the mitochondrial membrane potential in peripheral blood granulocytes from type 1 diabetic patients and their unaffected siblings using the mitochondrial indicator 5,5',6,6'-tetra chloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide (JC-1) in conjunction with flow cytometry (Matteucci et al., 2011). This was the first study to examine mitochondrial membrane potential of circulating leukocytes in type 1 diabetes families and to document consistent evidence for mitochondrial hyperpolarisation that was highest in type 1 diabetic patients and intermediate in their siblings. Fasting plasma glucose was the only correlate of leukocyte mitochondrial membrane potential. Confirming previous observations in type 1 diabetes families, siblings had fasting plasma glucose slightly higher than control subjects yet lower HbA1c levels. The combination of higher mean fasting plasma glucose, lower homeostasis model assessment of insulin sensitivity (HOMA-IS) and lower HbA1c levels suggested that siblings had both impaired basal glucose clearance rate and enhanced

We hypothesised that in type 1 diabetes families, radical-induced mitochondrial membrane potential oscillations may be synchronized toward polarized states. The positive association between mitochondrial membrane potential oscillations and fasting plasma glucose within the range from normal to dysglycemic conditions suggested that hyperglycaemic challenge implied increased glucose metabolisation, enhanced oxidant formation and

It is noteworthy that succination of proteins, which is an irreversible chemical modification of cysteine by the Krebs cycle intermediate fumarate, is increased by hyperpolarisation of

patients differed from age-matched control subjects in the higher levels of Lipoprotein (a) (Matteucci et al., 2000a). In the same study, we investigated the redox status and antioxidant defences in these families.

The premises were the following:


We suggested an alternative hypothesis, i.e. that oxidative stress preceded diabetes mellitus. In the case, indirect evidence for increased oxidative stress could be also detectable in nondiabetic relatives of type 1 diabetic patients. In order to provide evidence of a familial imbalance between radical production and antioxidant defences, we investigated indices of glucose and lipid metabolism, markers of plasma and cell lipid peroxidation, a novel marker of oxidant-induced protein damage, and the effects of oxygen radicals on erythrocytes of patients with type 1 diabetes and their relatives. We measured blood creatinine, glucose, HbA1c, cholesterol, triglycerides, Lipoprotein (a), fibrinogen, malondialdehyde, and advanced oxidation protein products. Erythrocyte response to oxidative stress (3-hincubation at 37°C with or without a radical generating system) was evaluated by measuring erythrocyte glutathione, erythrocyte malodialdehyde, and haemolysis. Plasma and erythrocyte malodialdehyde were found to be significantly elevated in diabetics and relatives than in controls. Basal erythrocyte glutathione was lower in diabetics and incubations of cells caused in diabetics a decrease in erythrocyte glutathione of lesser degree than in control subjects, while a significant increase in haemolysis. Among relatives, haemolysis was increased both at baseline and after incubation. Plasma malodialdehyde was associated with blood glucose, creatinine, and fibrinogen; basal erythrocyte malodialdehyde with plasma Lipoprotein (a), fibrinogen, and plasma malodialdehyde. Basal erythrocyte glutathione content correlated with serum glucose and erythrocyte malodialdehyde production.

In that occasion, we were pioneers of the research on redox balance in type 1 diabetes families. We presented first evidence that markers of lipoprotein metabolism (Lipoprotein (a)), oxidative stress (plasma and erythrocyte malodialdehyde), and cellular fragility (haemolysis) are abnormal in non diabetic relatives of type 1 diabetics supporting the view that familial elements even precede diabetes. It seemed reasonable that the same biologic markers considered major predictors of cardiovascular disease could also trace familial susceptibility to type 1 diabetes, just as they have been associated with the development of type 2 diabetes (Matteucci et al., 2000a).

Based on the finding of elevated circulating markers of lipid peroxidation and increased cellular fragility, we decided to complete and integrate our investigation with further biochemical measurements of possible first-chain initiating or stimulating factors in order to evaluate, in the same families, the contribution of extracellular antioxidants to the increased oxidative stress. We also aimed to understand the eventual relationship between oxidative stress and the abnormal sodium/hydrogen exchange activity previously observed (Matteucci et al., 2001). We were unable to find out any abnormalities in circulating metal ions (such as iron, transferrin, ferritin, copper, and ceruloplasmin) or extracellular antioxidant defences (such as serum uric acid, albumin, bilirubin,) that could favour oxidative stress in non-diabetic relatives of type 1 patients. On the contrary, we confirmed our previous finding of a generalised increase in sodium/hydrogen exchange activity. The rate of amiloride-sensitive hydrogen efflux from erythrocytes was significantly associated with both erythrocyte glutathione content and some markers of radical-induced damage such as plasma advanced oxidation protein products and malondialdehyde, erythrocyte osmotic fragility, and erythrocyte malondialdehyde accumulation under oxidative stress. Hence, this additional study provided the first *in vivo* demonstration of a significant association between oxidative stress and sodium/hydrogen exchange upregulation. The familiarly overactive sodium/hydrogen exchange itself could be viewed as further evidence pointing to the presence in these families of a redox disequilibrium where oxidation seems to be prevailing.

Taken into account that:

314 Type 1 Diabetes – Complications, Pathogenesis, and Alternative Treatments

patients differed from age-matched control subjects in the higher levels of Lipoprotein (a) (Matteucci et al., 2000a). In the same study, we investigated the redox status and antioxidant

 enhanced levels of free radicals found in diabetes mellitus and impaired glucose tolerance has long been assumed to be related to chronically elevated glucose levels

 oxidative stress was suggested to play a primary role in the pathogenesis of diabetes and its complications but Authors still discussed whether oxidation preceded the appearance of complications or it merely reflected their presence (Baynes and

We suggested an alternative hypothesis, i.e. that oxidative stress preceded diabetes mellitus. In the case, indirect evidence for increased oxidative stress could be also detectable in nondiabetic relatives of type 1 diabetic patients. In order to provide evidence of a familial imbalance between radical production and antioxidant defences, we investigated indices of glucose and lipid metabolism, markers of plasma and cell lipid peroxidation, a novel marker of oxidant-induced protein damage, and the effects of oxygen radicals on erythrocytes of patients with type 1 diabetes and their relatives. We measured blood creatinine, glucose, HbA1c, cholesterol, triglycerides, Lipoprotein (a), fibrinogen, malondialdehyde, and advanced oxidation protein products. Erythrocyte response to oxidative stress (3-hincubation at 37°C with or without a radical generating system) was evaluated by measuring erythrocyte glutathione, erythrocyte malodialdehyde, and haemolysis. Plasma and erythrocyte malodialdehyde were found to be significantly elevated in diabetics and relatives than in controls. Basal erythrocyte glutathione was lower in diabetics and incubations of cells caused in diabetics a decrease in erythrocyte glutathione of lesser degree than in control subjects, while a significant increase in haemolysis. Among relatives, haemolysis was increased both at baseline and after incubation. Plasma malodialdehyde was associated with blood glucose, creatinine, and fibrinogen; basal erythrocyte malodialdehyde with plasma Lipoprotein (a), fibrinogen, and plasma malodialdehyde. Basal erythrocyte glutathione content correlated with serum glucose and erythrocyte

In that occasion, we were pioneers of the research on redox balance in type 1 diabetes families. We presented first evidence that markers of lipoprotein metabolism (Lipoprotein (a)), oxidative stress (plasma and erythrocyte malodialdehyde), and cellular fragility (haemolysis) are abnormal in non diabetic relatives of type 1 diabetics supporting the view that familial elements even precede diabetes. It seemed reasonable that the same biologic markers considered major predictors of cardiovascular disease could also trace familial susceptibility to type 1 diabetes, just as they have been associated with the development of

Based on the finding of elevated circulating markers of lipid peroxidation and increased cellular fragility, we decided to complete and integrate our investigation with further biochemical measurements of possible first-chain initiating or stimulating factors in order to evaluate, in the same families, the contribution of extracellular antioxidants to the increased oxidative stress. We also aimed to understand the eventual relationship between oxidative stress and the abnormal sodium/hydrogen exchange activity previously observed

(Baynes and Thorpe, 1999; Vijayalingam et al., 1996),

defences in these families. The premises were the following:

Thorpe, 1999).

malodialdehyde production.

type 2 diabetes (Matteucci et al., 2000a).


in the last step of our research we measured the mitochondrial membrane potential in peripheral blood granulocytes from type 1 diabetic patients and their unaffected siblings using the mitochondrial indicator 5,5',6,6'-tetra chloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide (JC-1) in conjunction with flow cytometry (Matteucci et al., 2011). This was the first study to examine mitochondrial membrane potential of circulating leukocytes in type 1 diabetes families and to document consistent evidence for mitochondrial hyperpolarisation that was highest in type 1 diabetic patients and intermediate in their siblings. Fasting plasma glucose was the only correlate of leukocyte mitochondrial membrane potential. Confirming previous observations in type 1 diabetes families, siblings had fasting plasma glucose slightly higher than control subjects yet lower HbA1c levels. The combination of higher mean fasting plasma glucose, lower homeostasis model assessment of insulin sensitivity (HOMA-IS) and lower HbA1c levels suggested that siblings had both impaired basal glucose clearance rate and enhanced insulin-stimulated muscle glucose disposal.

We hypothesised that in type 1 diabetes families, radical-induced mitochondrial membrane potential oscillations may be synchronized toward polarized states. The positive association between mitochondrial membrane potential oscillations and fasting plasma glucose within the range from normal to dysglycemic conditions suggested that hyperglycaemic challenge implied increased glucose metabolisation, enhanced oxidant formation and hyperpolarisation of the mitochondrial membrane.

It is noteworthy that succination of proteins, which is an irreversible chemical modification of cysteine by the Krebs cycle intermediate fumarate, is increased by hyperpolarisation of

The Enlarging List of Phenotypic Characteristics That

association with the number of CD4+CD26-

**8. Concluding remarks** 

CD4+CD26-

Might Allow the Clinical Identification of Families at Risk for Type 1 Diabetes 317

However, the fluorescence intensity of CD26 expression on CD8+ lymphocytes revealed a significant decrease in type 1 diabetic patients compared with control subjects. Mean fluorescence of CD8+CD26+ cells was inversely correlated with the absolute number of

among CD8+ T cells in type 1 diabetes) as indicating a defect in successfully developed longterm memory CD8+ T cells or in CD8+ T cells activation, even though the negative

peripheral T cells. We intend to continue research in this field in consideration of the

Today, there is a great need to integrate molecular biology with whole organ physiology. Findings from molecular and cellular studies must be brought back to intact organ systems without loosing the physiological context (Königshoff et al., 2011). This is especially true in the field of metabolic diseases where the study of individual proteins and signalling pathways in detail may not be easily translated to the intact organism. Taken into account the enlarging list of phenotypic characteristics that might allow the early clinical identification of families possibly at risk for sporadic cases of type 1 diabetes, many

Fig. 4. Some of the potential mechanisms linking metabolic syndrome and T cell maintenance. First question: may insulin-resistance be the common denominator of the observed familial peculiarities? And therefore, second question: could an early correction of one/some of

immunomodulating role of the multifunctional CD26 (Ohnuma et al., 2011).

questions await an answer. We suggest the two main (in our opinion) issues.

cells (Matteucci et al., 2010). We interpreted the finding (low expression of CD26

T cell does not support a recent activation of

the inner mitochondrial membrane and develops in concert with mitochondrial and oxidative stress in diabetes (Frizzell et al., 2011).

### **7. Immunological functions in type 1 diabetes families**

Although type 1 diabetes is a T-cell–mediated autoimmune disease, until a few years ago relatively few studies have attempted to associate T-cell autoreactivity with disease progression, in comparison with efforts directed on monitoring autoantibodies, and those that have been performed were largely limited to CD4 T-cells (Roep, 2008). Currently, islet epitope-specific CD8 T cells are believed to have a pivotal role in the destruction process. Unfortunately, monitoring multiple epitope-specific CD8 T cell populations poses many technical problems. Recently, monitoring of CD8 T cells reactive to beta-cell-derived antigens has been performed using the combinatorial quantum dot technique, which has been validated using peripheral blood cells from recent-onset type 1 diabetic patients, their siblings, and control subjects (Velthuis et al., 2010). Moreover, during the progression of autoimmune diabetes, memory autoreactive regulatory CD8 T cells can be expanded that could effectively suppress the expansion of dominant and subdominant effectors (Khadra et al., 2010). Increasing evidence shows the significance of CD4 and CD8 regulatory T cells, expressing the marker CD25 or IL-2 receptor, in autoimmune disease models. On the contrary, very few study have dealt with the role of CD23 or low affinity IgE receptor. In 2004, given that abnormalities in redox balance clustered in type 1 diabetes families and the intracellular redox status seemed to modulate immune function, we aimed to investigate the relationship between oxidative stress and immunologic features. We measured oxidative markers, serum pro-inflammatory cytokines, soluble cytokine receptors, and subsets of peripheral blood lymphocytes (by varying combinations of CD4, CD8, CD23, and CD25) from type 1 patients, low-risk (i.e. without underlying islet autoimmunity) non-diabetic first-degree relatives of diabetic patients, and healthy subjects (Matteucci et al., 2004a). In these families, protein and lipid oxidation was confirmed from reduced sulfhydryl groups, increased advanced oxidation protein products, increased plasma and erythrocyte malondialdehyde. Relatives had decreased counts of monocytes, of cells coexpressing CD23 and CD25, and of CD25+ cells in peripheral blood. Patients with type 1 diabetes had similar defects and, in addition, showed decreased counts of peripheral CD4+CD8+ lymphocytes and increased serum levels of soluble receptors for IL-6 and IL-2. This was the first demonstration of leukocyte abnormalities in low-risk T1DM relatives, also presenting signs of oxidative stress. Moreover, our study reported first evidence that the oxidative stress observed in type 1 diabetes families was correlated to immunological hallmarks suggestive of different immunoregulatory mechanisms. A crucial question remained open: did the alteration in immune functions follow the altered intracellular redox status or vice versa?

More recently, we have characterised CD26 expression of T cell subsets in patients with type 1 diabetes because 1) high expression of CD26 among CD8+ T cells has been suggested to be a marker of effective long-term memory T cell formation typical of acute resolved viral infections (Ibegbu et al., 2009), and 2) an increased risk of persistent viral infections, such as hepatitis C (HCV), was reported among diabetic patients (Lonardo et al., 2009).

No significant difference was seen in percentages or absolute numbers of CD4+CD26+, CD4+CD26-, CD8+CD26+, and CD8+CD26- between type 1 diabetes and control people. However, the fluorescence intensity of CD26 expression on CD8+ lymphocytes revealed a significant decrease in type 1 diabetic patients compared with control subjects. Mean fluorescence of CD8+CD26+ cells was inversely correlated with the absolute number of CD4+CD26 cells (Matteucci et al., 2010). We interpreted the finding (low expression of CD26 among CD8+ T cells in type 1 diabetes) as indicating a defect in successfully developed longterm memory CD8+ T cells or in CD8+ T cells activation, even though the negative association with the number of CD4+CD26- T cell does not support a recent activation of peripheral T cells. We intend to continue research in this field in consideration of the immunomodulating role of the multifunctional CD26 (Ohnuma et al., 2011).
