Metal-activated C-peptide Facilitates Glucose Clearance and the Release of a Nitric Oxide Stimulus via the GLUT1 Transporter

Abstract

Objective:

Proinsulin C-peptide has been implicated in reducing complications associated with diabetes and improving blood flow. We hypothesized that incubation of erythrocytes with C-peptide would improve the ability of these cells to release ATP, a stimulus of nitric oxide production.

Research Design and Methods:

Erythrocytes obtained from rabbits (n=11) and humans (healthy and those with type 2 diabetes, n=7) were incubated with C-peptide (in the absence and presence of Fe (II) and Cr (III)) and the resulting ATP release was measured via chemiluminescence. This release was also measured in the presence and absence of phloretin, an inhibitor of the glucose transporter GLUT1, and mannose, a glycolysis inhibitor. To determine glucose transport, ¹⁴C-labelled glucose was added to erythrocytes in the presence and absence of the C- peptide/metal complex and the aforementioned inhibitors.

Results:

The release of ATP from the erythrocytes of patients with diabetes increased from 64 nmol/l (± 13 nmol/l) to 260 nmol/l (± 39 nmol/l) upon incubation of the cells in C-peptide. The C-peptide activity was dependent upon binding to Fe (II), which was extended upon binding to Cr (III). The increase in ATP release from the erythrocytes is due to metal-activated C-peptide stimulation of glucose transfer into the erythrocytes via the GLUT1 transporter. In the presence of C-peptide complexed to Cr (III), the amount of glucose transferred into the erythrocyte increased by 31%.

Conclusions:

When complexed to Fe (II) or Cr (III), C-peptide has the ability to promote ATP release from erythrocytes. This release is due to an increase in glucose transport through the GLUT1 transporter.

Diabetes is often typified by a lack of insulin production or a resistance to insulin that is produced. The role of the insulin hormone is to facilitate glucose transport across cell membranes where it is then consumed or stored by the cell. Improper clearance of the glucose leads to hyperglycaemia and many associated complications including retinopathy, neuropathy, renal failure, and cardiovascular disease.

Proinsulin C-peptide is created in the pancreas as a part of insulin production. For many years, this 31 amino acid peptide was thought to have minimal biological activity.[1] In the 1990's, researchers reported that C-peptide was able to ameliorate complications associated with diabetes in rat models including vascular and neural dysfunction, glucose utilization, and renal function.[2-4] Following these reports were others demonstrating the ability of C-peptide to improve blood flow in the microvasculature[5] and skin[6] and to improve the deformability of erythrocytes.[7] C-peptide has also been reported to increase eNOS activity.[8]

The work involving C-peptide with erythrocytes and eNOS activation becomes significant when considering the work by Sprague, which demonstrated the ability of erythrocytes to release adenosine triphosphate (ATP) in response to stimuli including hypoxia, deformation, and other agonists.[9-11] Importantly, ATP is a recognized stimulus of endothelium-derived nitric oxide (NO), a potent vasodilator. Sprague has demonstrated that this RBC-derived ATP is a determinant in vascular resistance in the isolated rabbit lung.[12]

Collectively, reports demonstrating the ability of C-peptide to increase blood flow in the microvasculature and glucose transport in certain cell types, and that ATP can be released from erythrocytes in response to certain stimuli, suggest a potential novel role for C-peptide in vivo. Therefore, we have performed studies to determine if the ability of C-peptide to mediate the production of endothelium- or platelet-derived NO is through an increase in the concentration of ATP released from erythrocytes subjected to incubation with physiological levels of C-peptide. Subsequent studies were performed to determine if the increase in ATP release was due to activation of the major glucose transporter found in the RBC (GLUT1). An increase in GLUT1 activity would enhance glycolysis, the primary route by which RBCs produce ATP, thus activating a reported c-AMP-dependent pathway by which ATP is released from the RBC.[13]

Here, we demonstrate that RBCs are indeed able to release increased concentrations of ATP for a period of about 24 h when incubated with C-peptide bound to Fe (II), whereas binding of C-peptide to Cr (III) results in sustained ATP release over a period of 3-5 days. A thorough examination indicates that the ATP release is related to an increase in glucose transport across the RBC membrane. Thus, C-peptide, when complexed to a metal, has properties similar to insulin by facilitating glucose transport across RBCs. Insulin does not perform this function directly on the major cellular component of the bloodstream (the RBC), thereby rendering the activity of the metal activated C-peptide crucial to proper blood glucose maintenance.

Subjects, Materials and Methods

Preparation of RBCs

Rabbits (New Zealand whites, males, 2.0-2.5 kg) were anesthetized with ketamine (8 ml/kg, i.m.) and xylazine (1 mg/kg, i.m.) followed by pentobarbital sodium (15 mg/kg i.v.). A cannula was placed in the trachea and the animals were ventilated with room air. A catheter was placed into a carotid artery for administration of heparin and for phlebotomy. After heparin (500 units, i.v.), the animals were exsanguinated. Human blood was obtained by venipuncture without the use of a tourniquet (antecubital fossa) and collected into a heparinized syringe. Blood was centrifuged at 500 × g at 4°C for 10 min. The plasma and buffy coat were discarded. The RBCs were resuspended and washed three times in a physiological salt solution [PSS; in mmol/l, 4.7 KCl, 2.0 CaCl₂, 140.5 NaCl 12 MgSO₄, 21.0 tris(hydroxymethyl)aminomethane, 11.1 dextrose with 5% bovine serum albumin (final pH 7.4)]. Cells were prepared on the day of use and experiments were finished within 8 h of removal from the animal or human subjects. All procedures were approved by the Animal Investigation Committee or the Human Investigation Committee at Wayne State University.

Measurement of ATP

Human C-peptide (American Peptide Co., Sunnyvale, CA), 0.25 mg (MW = 3020 g/mol), was dissolved in 100 ml of purified water (18.2 megaohm) to yield a concentration of 83 μmol/l. Next, an appropriate volume of this C-peptide solution was added to 10 ml of a 7% solution of RBCs to create a solution containing the C-peptide at concentrations in the 1-10 nmol/l range. Detection of the ATP release from the RBCs was performed using the luciferase assay and chemiluminescence measurement by a PMT. The resultant current from the PMT, which is proportional to the ATP-induced chemiluminescence, was measured as a potential by a data acquisition board controlled by a program written with the LabView software package (National Instruments, Austin, TX).

To ensure that cell lysis was not occurring during the assay, a solution of 0.01 mol/l glibenclamide was prepared by adding 49 mg of glibenclamide (Sigma-Aldrich, St. Louis, MO) to 2 ml of a 0.1 mol/l solution of sodium hydroxide. To this, 7.94 ml of a dextrose solution (1 g dextrose in 20 ml of purified water) was added. The mixture was heated carefully to 52 °C until all of the glibenclamide was dissolved. Once the solute was completely dissolved, 1 ml of this solution was added to 9 ml of PSS, resulting in a solution with a concentration 1 mmol/l. From this diluted solution, 2.5 ml were added to 2.5 mL of 7% hematocrit RBC solution, resulting in a 3.5% hematocrit solution of RBCs. This solution was allowed to incubate for 15 min. As a comparison, 2.5 ml of wash buffer without glibenclamide was added to 2.5 ml of 7% hematocrit RBCs. After 15 min, the RBC solutions were measured using the same methods as described above.

The solutions of Cr (III) and Fe (II) (derived from chromic chloride hexahydrate and ferrous ammonium sulfate hexahydrate, respectively) were prepared in purified water. The metal solutions were then added in equimolar amounts to the C-peptide solution (also in purified water) through a series of dilutions. Prior to addition to the RBCs, the metal-C-peptide mixture was diluted in PSS to avoid cell lysis that may occur between direct contact of RBCs with a non-buffered aqueous solution.

To determine the ability to inhibit glycolysis using mannose, 500 μl of a 5.5 mmol/l solution of D-mannose, that had been prepared in the PSS, was added to the 7% hematocrit RBCs solutions with and without the 1 nmol/l C-peptide. The resultant ATP release was measured after 6 h using the chemiluminescence assay described above. For the GLUT1 inhibition studies, the GLUT1 inhibitor phloretin was prepared by dissolving 0.100 g of phloretin (Sigma-Aldrich) in an appropriate amount of dimethyl sulfoxide, followed by dilution to 50.0 ml in phosphate-buffered saline yielding a final concentration of 0.5 mmol/l. Next, 1.5 ml of the 0.5 mmol/l phloretin solution was added to PSS containing C-peptide; after the addition of 500 μl of RBCs (70% hematocrit) to the phloretin/peptide mixture, the solution was diluted to a final volume of 5 ml, resulting in a solution containing a 7% hematocrit of RBCs in the presence of 10 nmol/l C-peptide and Cr (III). This solution was allowed to incubate for a period of 6 h, followed by measurement of the ATP release from the RBCs using the luciferase assay for ATP.

Mass Spectrometry Analysis

All experiments were performed using a Thermo model LTQ linear ion trap mass spectrometer (San Jose, CA). Samples were prepared by dissolving synthetic C-peptide in purified water at a concentration of 8.3 μmol/l. For metal binding studies, 3.4 μl of a 1 mmol/l solution of ferrous ammonium sulfate, ferric chloride or chromic chloride prepared in purified water were individually added to the C-peptide solution then introduced to the mass spectrometer at a flow rate of 0.5 μl/min by nanoelectrospray ionization (nanoESI). The spray voltage was maintained at 2.0 kV. The heated capillary temperature was 250°C. Mass spectra were acquired using the ‘enhanced’ (Figs. 3a, 3c and 3e) and ‘ultrazoom’ (Figs. 3b, 3d and 3f) resonance ejection scan modes of the linear ion trap mass spectrometer.

Fig. 3.

nanoESI mass spectrometry analysis of C-peptide and its metal adducts. (a) Mass spectrum of freshly prepared C-peptide, (b) High resolution mass spectrum of the [M+3H⁺]³⁺ region from panel a, (c) Mass spectrum of C-peptide incubated with iron (II) (d) High resolution mass spectrum of the [M+H⁺+Fe²⁺]³⁺ region from panel c, (e) Mass spectrum of C-peptide incubated with chomium (III), and (f) High resolution mass spectrum of the [M+H⁺+Cr³⁺]⁴⁺ region from panel e.

¹⁴C-labelled Glucose Studies

Experiments were performed using RBCs that had been washed in a low glucose concentration (0.55 mmol/l) PSS. The solution preparation for C-peptide and all inhibitors remained the same as for previous studies. Additionally, 0.042 μCi of ¹⁴C-labelled glucose was added to the PSS. This created a ratio of ¹⁴C-labelled glucose to non-radiolabelled glucose of 1:10. RBCs were added creating a 7% hematocrit solution and allowed to incubate for a period of 4 h. After centrifugation, the RBCs were combined with a scintillation cocktail and the radioactivity was measured using a scintillation counter over a period of 3 min for each sample.

Results

C-peptide-induced Release of ATP

The ATP released by RBCs was measured using an established chemiluminescence assay.[14, 15] Another aliquot from the same RBC sample was incubated in 1 nmol/l C-peptide and the resultant ATP release from the RBCs was measured every 2 h for a period up to 8 h. Figure. 1, which shows the normalized values of ATP released from the RBCs of n = 11 rabbits in the presence and absence of C-peptide, indicates that the addition of C-peptide resulted in an increased release of ATP from the RBCs. Importantly, it was later verified through mass spectrometric analysis that the C-peptide used to obtain the data shown in Fig. 1 was bound to trace levels of Fe (II), an impurity in the commercially available product. The increase in RBC-derived ATP observed over the 8 h time period was almost three times greater than that of the RBCs incubated with a control (buffer without C-peptide). The increase in measured extracellular ATP release was not due to cell lysis, based on the inhibition of ATP release when the RBCs were incubated in glibenclamide. Glibenclamide is classically recognized as a chloride channel inhibitor that is also known to block the release of ATP via its ability to inhibit the cystic fibrosis transmembrane regulator (CFTR) protein.[11]

Fig. 1.

Determination of ATP release from rabbit RBCs. The data shown are normalized values from the RBCs of n = 11 rabbits incubated in the presence and absence of 1 nmol/l C-peptide. As shown, the ATP release (determined by a chemiluminescence assay) from those cells incubated in the C-peptide (clear bars) increased significantly after 4 h (p < 0.005) and approximately 2.9 times over a period of 8h compared to RBCs incubated with no C-peptide (black bars). Although no metal was added, the C-peptide was bound to Fe (II) as verified by mass spectrometry. The metal was present as an impurity upon dissolution in the solvent (water). Error bars are ± SEM.

Restoration of ATP Release from the RBCs of Patients with Diabetes

Recently, it has been reported that the ATP released from RBCs obtained from the whole blood of patients with type 2 diabetes mellitus is approximately 50% of that released from the RBCs of healthy control patients.[14] It was concluded that the RBCs of the patients with diabetes may have released less ATP due to oxidative stress within the RBCs leading to a less deformable cell. A decrease in RBC deformability is a recognized trait of the RBCs obtained from patients with diabetes.[16, 17] Moreover, it has recently been reported that a decreased release of ATP from diabetic erythrocytes may be due to an inactivation of the G_i protein subunit.[13] Due to the ability of the C-peptide to increase ATP release from the RBCs of healthy rabbits, it was anticipated that the C-peptide may be able to increase the ATP release from the RBCs of patients with type 2 diabetes. The data in Fig. 2 show that, not only does the C-peptide have the ability to increase the ATP released (64 ± 13 nmol/l ATP at 0 h) from the RBCs of patients with type 2 diabetes, but it also has the ability to restore ATP levels (260 ± 39 nmol/l at 6 h, n = 7) to a value that is statistically equivalent to that of healthy control patients.[14] Also in Fig. 2 is data resulting from the analysis of RBCs obtained from healthy control patients. The ATP release from these subjects (260 ± 60 nmol/l, n = 7) in the presence of the peptide at 0 h was almost identical to that of the RBCs from the patients with type 2 diabetes after incubation with the peptide. Interestingly, the ATP release from the RBCs obtained from healthy control patients also increased (480 ± 109 nmol/l, n = 7) after 6 h of incubation with the C-peptide.

Fig. 2.

RBCs obtained from human patients with diabetes (grey bars) released 64 ± 13 nmol/l ATP at 0 h. After incubation with C-peptide for 6 h, the ATP release increased significantly (p < 0.005) to 260 ± 39 nmol/l. The release of ATP from RBCs obtained from human control patients (black bars) was also determined at 0 h (260 ± 60 nmol/l) and at 6 h (480 ± 109 nmol/l), also a statistically significant increase (p < 0.005). Interestingly, the C-peptide/metal adduct increased the ATP release from the RBCs obtained from patients with type 2 diabetes to a value that is statistically identical to the release of ATP from the RBCs of the healthy controls. Error bars are ± SEM (n = 7).

Mass Spectrometry Analysis of Peptide-Metal Interactions

Interestingly, the C-peptide exhibited significantly reduced activity, with respect to its ability to increase the concentrations of RBC-derived ATP, after a period of 24-36 h following its preparation in water. Analysis of the C-peptide by electrospray ionization mass spectrometry (ESI-MS) indicated that the peptide had not undergone any significant modification or degradation during this time, even after remaining in solution for periods > 30 days. In addition, there have been no reports indicating secondary structures of C-peptide. Figure 3a shows the mass spectrum obtained following analysis of an 8.3 μmol/l solution of C-peptide freshly prepared in water. A high resolution scan of the triply charged region of the spectrum from Fig. 3a revealed that the peptide had formed adducts with various ions including sodium and potassium (Fig. 3a). Interestingly, this mass spectrum also showed that the C-peptide had formed an adduct with Fe (II). The deliberate addition of Fe (II) to the fresh C-peptide solution at a molar ratio of 2:1 resulted in the formation of this adduct as an abundant species (Fig. 3c). A high resolution scan of the region of the mass spectrum containing the dominant triply charged [M+H⁺+Fe²⁺]³⁺ adduct is shown in Fig. 3d. In order to extend the activity of the C-peptide, it was thought that the addition of a metal ion that has a slower exchange rate with ligands in solution could be employed with more success towards maintenance of biological activity. Therefore, Cr (III) was added to a freshly prepared 8.3 μmol/l solution of C-peptide at a molar ratio of 2:1 and analyzed by ESI-MS (Fig. 3e). It can be seen in Figs. 3e and 3f that the dominant adduct produced by this addition corresponded to the monomeric [M+H⁺+Cr³⁺]⁴⁺ ion, with a non-covalent dimeric C-peptide adduct corresponding to [2M+Cr³⁺.6H₂O+2H⁺]⁵⁺ also being observed, albeit at a relative abundance of only 35% of that of the [M+H⁺+Cr³⁺]⁴⁺ ion.

Extension of C-peptide Activity though Binding to Cr(III)

Based on the data shown in Fig. 3, which demonstrated the ability of the C-peptide to bind to Fe (II) and Cr (III), RBCs were incubated in C-peptide solutions containing Fe (II) or Cr (III) and their subsequent ability to release ATP was determined (Fig. 4). As a negative control, RBCs were incubated in C-peptide that had been kept at 4 °C for >30 days. As expected, this solution no longer had the ability to induce ATP release from the RBCs. Moreover, C-peptide that had been purified using HPLC also demonstrated no activity in the presence of RBCs. Furthermore, RBCs incubated with the metal ions in the absence of C-peptide also did not result in any increase in ATP release. In contrast, an increase in ATP release was observed after incubation of the RBCs for 6 h in a 1 nmol/l solution of the inactive C-peptide to which 1 nmol/l Fe (II) or Cr (III) had been added, clearly demonstrating that the activity of the C-peptide was restored upon metal binding.

Fig. 4.

Normalized ATP release from fresh RBCs after incubation with a C-peptide solution (P) in contact with a metal after 6 and 72 h. It is clear that at 6 h the Fe (II) (clear bars) and Cr (III) (patterned bars) containing C-peptide solutions result in the same increased level (p < 0.005) of ATP release; however, after 72 h, the ATP release from RBCs incubated with the Fe (II) peptide solution decreased to a level similar to that of RBCs (black bars) in the absence of any peptide. The Cr (III) peptide solution maintained its level of activity. Error bars are ± SEM (n = 4). It should be noted that, although a very small increase, the difference in ATP release from the RBCs incubated in the 72 h Fe (II)/peptide solution is significant from the ATP release from the RBCs in the absence of any peptide (p < 0.01).

Although the Fe (II)-bound C-peptide had the ability to increase the ATP derived from the RBCs, its activity was observed to decrease after about 24 h. For example, the activity of the C-peptide solution 72 h after the addition of Fe (II) generally showed only a very slight increase from that of RBCs incubated in the absence of C-peptide (Fig. 4). Deliberate oxidation of a fresh Fe(II)-bound C-peptide solution, or the direct addition of Fe (III) to a fresh C-peptide solution, followed by mass spectrometry analysis, revealed only minimal formation of an Fe (III) adduct, suggesting that oxidation of Fe (II) to Fe (III) was responsible for this loss of activity (data not shown). Furthermore, RBCs incubated with a solution of C-peptide to which Fe (III) had been freshly added resulted in no increase in ATP release from these cells.

In contrast to that observed for the C-peptide in the presence of Fe (II), the increase in RBC-derived ATP release for the C-peptide solution 72 h after the addition of Cr (III) was observed to be essentially the same as that for the 6 h time period, indicating that the activity of the C-peptide could be extended to at least 3 days upon binding to Cr (III).

Enhanced Glucose Transport

In separate reports, Sprague has suggested cellular pathways explaining the mechanism for the release of ATP from RBCs.(16) While there have been subtle differences, each mechanism reported involves activation of adenylyl cyclase and subsequent production of c-AMP from ATP. Glycolysis is the main, if not only route, by which RBCs produce ATP. Therefore, experiments were performed to determine if the increased ATP release from the RBC in the presence of the C-peptide-metal complex was due to enhanced c-AMP production (via an increase in glycolytic ATP, a substrate for adenylyl cyclase). To test this hypothesis, RBCs were incubated in the presence and absence of phloretin, a GLUT1 inhibitor, and C-peptide activated with Cr (III). As shown in Fig. 5a, the Cr (III)-activated C-peptide resulted in an increase in ATP release of 74%. This increase was not obtained when the RBCs were incubated with phloretin prior to introduction of the Cr (III)-activated C-peptide. These results suggest that the ATP release from the RBCs is dependent upon the ability of the C-peptide to increase glycolysis within the RBC via increased cellular glucose transport though the GLUT1 transporter. However, phloretin is a chloride ion channel inhibitor and, therefore, may also inhibit ATP release from the RBCs. Additionally, D-mannose inhibits RBC glycolysis most likely though the glycolytic enzyme phosphoglucose isomerase. Upon addition of a mannose solution to RBCs and 1 nmol/l C-peptide, the ATP release was reduced to levels of RBCs alone, while the RBCs with metal-activated C-peptide saw increased levels of ATP as shown in Fig. 5b.

Fig. 5.

In (a) RBCs incubated with 10 nmol/l Cr (III)-activated C-peptide [P+Cr (III)] showed a significant (p < 0.005) increase in ATP release of approximately 74%. However, when RBCs were incubated with the GLUT1 inhibitor phloretin (PR) prior to addition of [P+Cr (III)], the ATP release was only about 63% of the value for the RBCs alone. Error bars are ± SEM (n = 8). In (b) the ATP release of RBCs that had been incubated with mannose (M) and [P+Cr (III)] showed a 57% decrease in ATP release over RBCs incubated with [P+Cr (III)] alone, returning the ATP release to a level that is not significantly different from RBCs. However, the RBCs incubated with [P+Cr (III)] resulted in a signal intensity that was approximately 64% higher than RBCs alone, constituting a significant increase (p < 0.005) Error bars are ± SEM (n = 5).

To provide further evidence that the metal-activated C-peptide facilitates glucose transport into the RBC, ¹⁴C-labelled glucose was included in the PSS at a 1:10 ratio with unlabelled glucose creating a competition between the different glucose isotopes. As shown in Fig. 6, the amount of glucose entering the RBC increased by 31% when the RBC solution contained Cr (III)-activated C-peptide. While this increase in glucose transport was not as high as the ATP released from the RBCs, it should be noted that the ratio of radio-labelled glucose to unlabelled glucose was not stoichiometrically equivalent. In addition, RBCs incubated with the metal ions in the absence of C-peptide did not result in any increase in glucose transport across the RBC membrane, or as mentioned above, RBC-derived ATP release.

Fig. 6.

RBCs incubated with 0.042 µCi ¹⁴C labelled glucose in a solution of PSS containing normal glucose. After 4 h, the RBCs incubated with 10 nmol/l Cr (III)-activated C-peptide [P+ Cr (III)] showed approximately 31% increase in radioactive counts per minute (CPM), while RBCs incubated with only C-peptide (P), Cr (III)-activated C-peptide plus the GLUT1 inhibitor phloretin [P+ Cr (III) + PR], or only phloretin (PR) all showed no increase in CPM. Error bars are ± SEM (n = 4).

Discussion

The results presented here show that C-peptide has the ability to increase the release of ATP from erythrocytes due to an activation of the GLUT1 transporter. Importantly, the activity is only present when the C-peptide is bound to a metal such as Fe (II) or Cr (III). Recent reports have provided evidence suggesting a biological role for C-peptide and many of these positive aspects of C-peptide have been debated by Wahren. [18] For example, it has been established that C-peptide increases renal function in patients with type 1 diabetes,[19] endoneurial blood flow,[20] and blood flow in the circulation.[6, 21]

The improvement in blood flow reported by different groups applying C-peptide may be explained by the increase in ATP release from erythrocytes that are subjected to the C-peptide. ATP is a recognized stimulus of NO production in the endothelium and the ability of the C-peptide to release increased levels of ATP from the RBCs is verified by the data in Fig. 1. It is important to note that in Fig. 1, the peptide did have trace levels of metal bound to it in the form of Fe (II). The presence of this metal, which occurred as a trace level impurity in the commercially available product, was verified by the mass spectrometry experiments. As stated above, in the absence of metals such as Fe (II) or Cr (III), the C-peptide's activity (in terms of its ability to facilitate glucose transport and stimulate ATP release from the RBCs) was not present.

The data presented in Fig. 3 demonstrates that C-peptide has the ability to bind metals. In addition, the results shown in Fig. 4 suggest that when C-peptide binds to Fe (II) there is activity that lasts for about 24 h. However, when binding to Cr (III), the activity lasts for at least 3 days. These results are interesting when one considers that the exchange rates for Fe (II) binding to most ligands is generally faster than the exchange rates for Cr (III) species binding to similar ligands. In other words, if the Fe (II) binds to the C-peptide, its activity lasts for a short period, but upon replacement of the Fe (II) (for example, with Na⁺ or K⁺), the C-peptide has no activity.

In a separate report,[22] Luzi has debated the negative concepts of C-peptide. For example, despite efforts, a receptor for C-peptide has never been reported. However, it is highly probable that receptor-based experiments on RBCs have yet to be performed. It is also highly likely that, if a receptor for C-peptide does exist, the C-peptide would need to assume a proper structure for receptor binding. C-peptide has previously been shown to lack a stable secondary structure [23,24]. One hypothesis to explain the results observed here is that coordination of Cr (III) to specific sites within the C-peptide molecule may result in the formation of a stable secondary structure that facilitates intracellular signal transduction via conformational dependent interactions between the C-peptide-metal complex and a specific (yet to be identified) cell surface receptor. Alternatively, coordination of Cr³⁺ metal ions (or other metal ions) may facilitate the formation of a stable bio-active C-peptide dimer. Further structural characterization studies are currently underway to examine these hypotheses. The need for metal activation may also help explain the lack of successful clinical trials involving C-peptide.

Another negative concern with C-peptide is that the question still remains why people with type 2 diabetes have many of the well-known complications associated with diabetes yet have elevated concentrations of C-peptide. Recently,[14] Carroll et al reported that the ATP released from the RBCs of patients with type 2 diabetes was approximately 50% of that ATP released from healthy controls. It was concluded from this study that a decrease in G6PD activity, a common trait in the diabetic erythrocyte, was leading to a less deformable cell and a decrease in G-protein activation upon deformation. Here, the ATP release from RBCs obtained from a sample set of patients with type 2 diabetes is shown to revert to normal levels in the presence of metal-activated C-peptide (Fig. 2). Interestingly, in a separate report, Sprague has demonstrated that decreased ATP levels were released from the RBCs of patients with type 2 diabetes due to improper accumulation of c-AMP levels in the erythocytes of these patients. Therefore, in such a scenario, C-peptide activation of glucose transport and subsequent ATP release from the RBC may not occur due to other complications within the diabetic erythrocyte itself, such as glycation of key proteins in the ATP release pathway.[13]

Future work required will involve the determination of the structure-function relationship between metal-activated C-peptide and RBCs. Specifically, a quantitative determination of those metals actually bound to the C-peptide in vivo will be imperative. For example, recent work has suggested that a polymorphism in the zinc transporter SLC30A8, which is expressed in β-cells where proinsulin is produced, has a direct correlation with type 2 diabetes. The work presented here may be a platform to explain the implications of this zinc transporter problem, [25] especially considering yet unpublished results in our labs indicating that Zn (II) also has the ability to stimulate the activity of the C-peptide upon adduct formation.