Search

Mitochondrial glycerol phosphate dehydrogenase (mGPD) is the rate-limiting enzyme of the glycerol phosphate redox shuttle. It was recently claimed that metformin, a first-line drug used for the treatment of type 2 diabetes, inhibits liver mGPD 30-50%, suppressing gluconeogenesis through a redox mechanism. Various factors cast doubt on this idea. Total-body knockout of mGPD in mice has adverse effects in several tissues where the mGPD level is high but has little or no effect in liver, where the mGPD level is the lowest of 10 tissues. Metformin has beneficial effects in humans in tissues with high levels of mGPD, such as pancreatic β-cells, where the mGPD level is much higher than that in liver. Insulin secretion in mGPD knockout mouse β-cells is normal because, like liver, β-cells possess the malate aspartate redox shuttle whose redox action is redundant to the glycerol phosphate shuttle. For these and other reasons, we used four different enzyme assays to reassess whether metformin inhibited mGPD. Metformin did not inhibit mGPD in homogenates or mitochondria from insulin cells or liver cells. If metformin actually inhibited mGPD, adverse effects in tissues where the level of mGPD is much higher than that in the liver could prevent the use of metformin as a diabetes medicine.

The involvement of Zn2+ in the inhibitory action of insulin and phenformin on bulk proteolysis was studied in the Langendorff rat heart with a Zn(2+)-buffering perfusate (0.1 mM citrate, physiological complete amino acids and 0.2% albumin). Proteins were biosynthetically labeled in vitro for 10 min with [3H]leucine. Rapidly degraded proteins were eliminated during a 3-h preliminary degradation without insulin or added Zn2+ (2 mM nonradioactive leucine). Insulin (5 nM), the lysosomal inhibitor chloroquine (30 μM), and the biguanide antihyperglycemic agent phenformin (2 μm) each caused a sustained 35–40% inhibition of [3H]leucine release beginning within 1–2 min and reaching a maximum at 1–1.5 h. When these agents were combined, their simultaneous proteolytic inhibitory effects were not appreciably > the effect of chloroquine alone. Infusion of supraphysiological perfusate Zn2+ (>15 μM) mimicked the inhibitory effect of insulin and chloroquine on lysosomal proteolysis. Infusion of supraphysiological Co2+, Mn2+, Fe2+, and Cr3+ (30 μM, 0.5 h) caused no change in proteolysis; however, 30 μM Cu2+ caused a slight inhibition. Presumptive chelation of the background (∼20 nM) Zn2+ by infusion of 3 μM CaNa2 EDTA caused no change in protein degradation over 1–2 h. The infusion of a physiological concentration of 1 or 5 microM Zn2+ (as ZnCl2) caused no change in protein degradation over 1–2 h. Biguanides are known to reversibly form a Zn2+ complex with affinity < that of Zn2+ for EDTA. Prior infusion of 3 μM CaNa2 EDTA inactivated the proteolytic inhibitory effect of maximal (2 μM) phenformin over at least 1.25 h of concurrent infusion. restoration of physiological 2 μM Zn2+ immediately restored the insulin-mimetic proteolytic inhibitory action of phenformin under continued 3 μM CaNa2 EDTA (5 μM total Zn2+, 2 μM nonchelated Zn2+). The time course and potency of submaximal insulin (

McN-3495 [chemical name: N-(1-methyl-2-pyrrolidinylidene)-N'-phenyl-1-pyrrolidinecarboximidamide], a new, oral, effective hypoglycemic agent that is structurally distinct from the biguanides, has been shown to be mechanistically distinct from the biguanides and to lack comparable side effects in animals. The biguanides (phenformin, buformin, metformin), which were active when glucose was given orally to fasted rats and dogs, failed to lower fasting blood glucose or improve glucose tolerance when the glucose was administered parenterally, thus suggesting that biguanides inhibit intestinal glucose absorption. McN-3495, however, lowered the fasting concentrations of glucose and increased glucose disappearance whether the glucose was administered orally or parenterally. Furthermore, intestinal glucose transport, measured by use of the everted sac technique, was inhibited after oral pretreatment of rats with buformin and phenformin but not with McN-3495. However, both phenformin and McN-3495 inhibited glucose transport when added directly in vitro, although McN-3495 did so at higher concentrations. The reasons for the differing actions of McN-3495 and the biguanides on intestinal transport in vivo and in vitro are discussed. New models for biguanide-induced lactacidemia using normal and streptozotocin-diabetic rats were developed. In addition, a number of the gastrointestinal side effects of phenformin use were observed in animal experiments. McN-3495, at doses above projected therapeutic doses, however, did not produce lactacidemia or comparable gastrointestinal side effects. Thus, McN-3495 is a novel oral hypoglycemic agent, probably safer than phenformin, that could provide better control of blood glucose in diabetic patients.

The potentiation of oral hypoglycemic drugs by the antilipemic agent halofenate is reported. Forty-seven diabetic patients were treated for 48 weeks with halofenate, clofibrate, or placebo. Five patients in the halofenate group were taking phenformin plus either chlorpropamide or tolbutamide. Their average initial fasting plasma glucose was 160 mg./dl. All five patients experienced a slow but substantial fall in fasting plasma glucose. The mean fasting plasma glucose for the five patients after 80 days of halofenate treatment was 63 mg./dl. As oral treatment for diabetes was reduced, the fasting plasma glucose returned to prehalofenate levels. In this study, we did not detect an effect of halofenate on the fasting plasma glucose of diabetic patients treated with insulin or on the fasting plasma glucose levels of patients treated with diet alone.

The effect of fixed doses of oral hypoglycemic agents and placebo (diet alone) on the blood glucose, serum insulin, triglyceride, and cholesterol responses during oral glucose tolerance tests done annually for up to four years' follow-up was studied, in a doubleblind manner, in five groups of mild male chemical diabetics. The drugs used were chlorpropamide (100 mg. O.D.), tolbutamide (500 ing. b.i.d.), phenformin (50 mg. O.D.), acetohexamide (250 mg. O.D.), and placebo. Each subject was given an individualized diet aimed at attaining and maintaining ideal weight. Comparison by chi-square analysis between the placebo group and each of the drug groups showed (a) no significant differences with regard to the number of subjects with normal glucose tolerance in each of the tests and (b) no change in the insulin secretion dynamics. Comparison between the initial test and each of the subsequent tests within each group showed (a) a greater number of subjects with normal glucose tolerance in the first follow-up test in the chlorpropamide group only, (b) no change in the insulin secretion dynamics except in the chlorpropamide group, where there was an increased insulin/glucose ratio in the first follow-up test, and (c) no change in the fasting serum triglyceride and cholesterol levels.

Gluconeogenesis from lactate and hepatic cell pH (pHi) were measured in the isolated perfused livers of starved guinea pigs in the presence and absence of phenformin (phenethylbiguanide). The observed decrease in lactate consumption and glucose output in the presence of phenformin was associated with a fall in pHi. The fall in glucose output observed was considerably greater than accountable for by the decrease in lactate consumption. A possible mechanism for the pathogenesis of clinical lactic acidosis due to phenformin therapy is suggested.

The kinetics of phenformin and its metabolite, p-hydroxyphenethylbiguanide, was studied in eight diabetic patients with varying degrees of renal impairment. Plasma and urinary phenformin and p-hydroxyphenethylbiguanide levels were determined by the multiple selected ion monitoring technique. Phenformin half-lives were unrelated to the degree of renal impairment, whereas reduced renal clearances of inulin and creatinine were significantly correlated with a prolonged half-life of the metabolite. The excretion of p-hydroxyphenethylbiguanide was quite variable (between 4.9% and 27% of total urinary drug loss), probably due to a genetic polymorphism of hepatic mechanisms for hydroxylation. A reduced formation of the metabolite was concomitant with marked increases in the amount of circulating phenformin. A positive reciprocal correlation was detected between areas under the plasma curve of phenformin and both the renal clearance of the unchanged drug and the percentage of metabolite formation. A reduced hydroxylation of phenformin seems, therefore, to be responsible for the high plasma levels of the drug previously described in toxic patients.

Eighteen diabetic patients with lactic acidosis (L.A.) were analyzed for possible causal factors, metabolic changes, and efficacy of treatment. An antecedent phenformin therapy was performed in fifteen cases and was associated with renal insufficiency in ten cases and liver disease in eight cases. Tissular anoxia of primary hemodynamic or respiratory origin was absent in all cases. The severe metabolic acidosis (pH = 6.93 ± 0,03 ; HCO3¯ = 6 ± 1 mM ; PaCO2 = 18 ± 2 mm. Hg) and hyperlactatemia (14.2 ± 0.3 mM) were associated with high lactate/pyruvate ratio (70 ± 22). High alanine levels (up to 4.6 mM) were measured in some of these patients. High beta-hydroxybutyrate levels were sometimes measured (up to 7.6 mM), and substantial amounts of acetoacetate were also detected in twelve cases. Glucagon level was always increased (1,050 ± 240 pg./ml.), and insulin/glucagon ratio was low. Cortisol (49 ± 10 µg./100 ml.) and HGH (10.8 ± 2.6 ng./ml.) were also elevated. Increased plasma levels of phenformin were measured in five L.A. diabetic subjects (50 ± 5 µ./ml.) by comparison with other phenformin-treated diabetic subjects. The specificity of the assay was investigated, and phenformin metabolites were characterized by thin-layer chromatography. For the treatment of L.A., adjunction of dialysis and furosemide improved the efficacy of early and massive sodium bicarbonate infusion. It is suggested that accumulation of phenformin via renal insufficiency plays a determinant role in causing L.A. through an impairment of lactate metabolism in the liver. An accelerated epuration of the drug may be helpful. in therapy of L.A. Phenformin treatment should be avoided in case of renal and/or liver insufficiency.

Body weights, blood glucose concentration, and therapeutic and clinical data were summarized by computer for each patient visit to our diabetes clinic between 1965 and 1970. Homeostasis of body weight was examined by plotting regression of body weight on time for patients making five or more visits to the clinic during the study period. A similar analysis was made of blood glucose values. Data from 131 patients treated with oral agents or by dietary regulation alone and forty-one patients treated with insulin were analyzed. Precision of regulation was assessed by calculating the coefficient of variation of body weight while the trend of body weight was assessed by calculating the slope of the linear regression equation relating body weight and time. In both treatment groups the individual slopes for body weight clustered closely around zero and the mean slopes did not differ significantly from zero. The coefficient of variation of body weight was also small, averaging 4.6 and 3.7 per cent of body weight in the oral agent and insulin treatment groups respectively. In contrast, blood glucose fluctuated widely in the two groups. Mean slopes for glucose versus time did not differ significantly from zero, however. The remarkable stability of body weight in this diabetic population despite widely varying modes of therapy, glucose concentrations, and degrees of body adiposity suggests that signals controlling long-term feeding behavior are either not related to these variables or, more likely, are multiple and sufficiently redundant to permit adaptation to perturbations of several metabolic variables.

At concentrations below 50 μM, phenethylbiguanide enhanced the initial rate of enertized Ca 2+ uptake into energized guinea pig liver mitochondria by as much as 45 per cent; Ca 2+ -stimulated O 2 uptake increased in parallel. The biguanide concentration that enhanced Ca 2+ uptake maximally was at least 15 times lower than that required for 50 per cent inhibition of respiration. Kinetic studies indicated that the enhanced rate of Ca 2+ transport resulted from an increase in V max , while Km for Ca 2+ was unaffected by the biguanide. Several other organic cations known to lower blood sugar in intact animals or to block the hepatic gluconeogenic response to glucagon also enhanced the mitochondrial Ca 2+ uptake rate; three of these compounds did not inhibit respiration even at high concentrations. One organic cation, triethyltin, which is a potent respiratory inhibitor that does not affect blood sugar, had no effect on Ca 2+ uptake. We concluded that enhancement of the mitochondrial Ca 2+ uptake rate is related to the mechanism of therapeutic blood sugar lowering by these drugs, probably by impairing the gluconeogenic response of the liver to glucagon.

Some products of the sorbitol metabolic pathway of glucose were extracted from peripheral nerves and spinal cords of rats, identified, and measured. Within a few days after a diabetic state was induced with streptozotocin, glucose, sorbitol, and fructose in nerve and cord had risen sharply as blood glucose increased. These levels were partially controlled when the blood sugar was restricted by insulin, but they increased rapidly when the insulin was withheld. Withdrawal of treatment resulted in a much greater rise of the fructose in the nervous system in animals that had been diabetic for six weeks before institution of treatment with insulin than in animals treated from induction. When phen-formin was used as a controlling agent in place of insulin, accumulation of glucose and fructose in the nervous system was enhanced, whereas fructose was found to decrease when tolbutamide was used. Earlier workers have shown that the sorbitol pathway seems to be associated with the Schwann cell and that dysfunction of this cell leads to segmental demyelination. We find that accumulation in nerves of the products of the sorbitol pathway may be minimized when blood sugar is controlled; since we have evidence that in human beings with diabetes good control improves the state of myelination of already damaged nerves, it is postulated that herein lies the mechanism whereby control of diabetes is likely to lead to a reduction in the incidence of nerve damage.

In rat adipose tissue, phenformin (5 × 10−4M) inhibited the oxidation of C1 and C6 of glucose to CO2, blocked the stimulatory effect of insulin (32 μU./ml.) on Ct oxidation, inhibited lipogenesis from C1 and C6 of glucose, and abolished the stimulatory effect of acetate on glucose oxidation and lipogenesis. These effects were accompanied by lactate accumulation. Such inhibitory effects suggest that phenformin interrupts the pentose shunt and lipogenesis, both of which are pyridine nucleotide-linked processes. Phenformin was shown, by inhibition of oxidation of pyruvate C1, to interfere with the sequence at the point of pyruvate decarboxylation, since acetate incorporation was not affected. Both pyruvate decarboxylation and oxidation of acetate in the tricarboxylic acid cycle were blocked, presumably by interference with electron transport. This effect of phenformin on pyruvate decarboxylation is sufficient to explain the observed reduction in pentose shunt activity and lipogenesis as well as the increased lactate accumulation.

The relationship between the biguanide inhibition of oxidations and their effect on glucose metabolism in vitro has been investigated. We found a good correlation between the concentration of biguanide causing a halfmaximal inhibition of pyruvate oxidation (Ki) and the concentration eliciting a half-maximal stimulation (Ks) of glucose uptake in several tissues from rat, guinea pig and pigeon. Guinea pig epididymal fat pad was the most sensitive tissue tested. Gluconeogenesis by minced guinea pig liver was inhibited by the biguanides. The ability of the biguanides to inhibit gluconeogenesis correlated well with their ability to inhibit pyruvate oxidation. Inhibition of oxidations by the biguanides does not appear to be a result of a direct inhibition of the enzymes catalyzing these oxidations but seems to be secondary to a primary effect on the coupled processes of electron transport and oxidative phosphorylation. We have summarized and discussed objections to the theory that the metabolic actions of the biguanides are a result of their inhibition of oxidative processes. Our conclusion is that the inhibition of oxidations caused by the biguanides is directly related to their metabolic and hypoglycemic effects.

We have studied the effect of sulfonylurea drugs on plasma free fatty acid and blood glucose concentrations in man. The concentrations of plasma free fatty acids and blood glucose were measured after the oral administration of a single dose of sulfonylurea drugs to fasting, maturity-onset, ketoacidosis-resistant diabetic patients. The changes in concentration of glucose and free fatty acids were not parallel as they are after systemic insulin or intravenous tolbutamide. There was no significant decrease in glucose concentration during the first hour after oral administration and little decrease at the second hour, but then a decrease to a minimum at five to six hours. During the first two hours, however, the concentration of free fatty acids decreased significantly despite little change in glucose concentration. At two hours, the concentration of free fatty acids decreased to its minimum, despite the fact that glucose concentration continued to decrease until five or six hours. These results are consistent with our earlier in vitro data and with the interpretation that the sulfonylurea drugs may have a direct antilipolytic effect.

The effect of phenformin on the uptake of glucose and water by isolated full-thickness sheets of human ileum has been studied. A significant decrease in mucosal glucose uptake was found, and significant glucose translocation to the serosal surface was abolished. This was accompanied by a proportional decrease in the production of lactate and in the net transport of water across the tissue. The magnitude of the observed changes does not support the hypothesis that overt glucose malabsorption is of significance in the therapeutic action of phenformin but confirms previous observations that the kinetics of absorption of a glucose load may be altered.

l-(2-pyrimidinyl)-4-imino-l, 4-dihydropyridine hydrochloride is a new compound which after oral administration of a single dose of 400 mg. causes significant reductions of hyperglycemia and insulin response in human subjects after an oral glucose load. These results were obtained when compared to a single dose of 200 mg. of phenformin hydrochloride or placebo in a double-blind completely randomized crossover study with placebo and phenformin. Both drugs also produced a significant fall in the fasting serum glucose value before the oral glucose load without any significant effect on serum insulin values. Both drugs produced similar adverse symptoms in the six healthy subjects who participated in the study.

The effect of phenformin on the rates of lactate and glucose turnover, interconversion, and oxidation was determined in five obese subjects with normal glucose tolerance. L (+) lactate-U (50 μc.) and glucose 1-C-14 were administered over four-hour periods on consecutive days by a primed infusion technic. One hundred to 150 mg. of phenformin was administered daily, and the turnover studies were repeated on the sixth and seventh days of therapy. Phenformin increased glucose turnover by 10 per cent, from 95.6 ± 4.4 to 106.2 ± 4.7 (mean ± S.E.M.) mg,/kg./ hr. and glucose recycling by 100 per cent, from 8.7 ± 1,4 to 19.3 ± 1.3 mg./kg./hr. In contrast, the changes in glucose oxidation were inconsistent, decreasing in three subjects and increasing in two. As expected, the blood glucose concentration and glucose pool were unchanged during treatment with phenformin. Lactate turnover increased 30 per cent, from 63.5 ± 3.5 to 82.8 ± 5.7 mg./kg./hr., and lactate derived from glucose (glucose conversion to lactate) increased 25 per cent, from 57.9 ± 5.5 to 71.8 ± 6.9 mg./kg./hr. Lactate conversion to glucose increased in all subjects by a mean of 13.1 ± 1.2 mg./kg./hr., from 15.2 ± 1.2 to 28.2 ± 1.8 mg./kg./hr. The increase in lactate conversion to glucose was almost identical to the increase in glucose turnover and glucose conversion to lactate. During therapy with phenformin, 34.1 ±1.2 per cent of the lactate turnover was converted to glucose in contrast to 22.1 ± 1.3 per cent before therapy. These results demonstrate that phenformin enhances the interconversion of glucose and lactate in subjects in whom it exerts no hypoglycemie effect.

McN-3495, a new compound unrelated structurally to the sulfonylureas or phenformin, has been found to produce a hypoglycemic effect in nondiabetic rats, dogs, mice, and monkeys. The minimum effective dose of McN-3495 that lowers fasting blood glucose and improves glucose tolerance was found to be about 2.5 to 5 mg. per kilogram, per os, except in fasted monkeys, in which a tenfold greater potency was observed. When McN-3495 was given repeatedly for three to five days, no tolerance to the hypoglycemic activity occurred and no changes in other biochemical parameters were observed. In addition to being three to four times more potent than tolbutamide, McN-3495 also differs from the sulfonylureas in lowering blood glucose concentrations of streptozotocin-diabetic rats and db/db mice, and, moreover, oral administration to normal fasted dogs did not produce the characteristic rise in insulin concentrations observed with tolbutamide. Furthermore, unlike the biguanides, McN-3495 can lower dog and rat fasting blood glucose concentrations and can improve glucose tolerance whether the glucose is administered orally or parenterally. However, McN-3495, as phenformin, fails to work in totally depancreatized dogs.

Streptozotocin-diabetic rats were treated with di-isopropylammonium dichloroacetate (DIPA) via an orogastric tube in doses of 25 or 50 mg./l kg. twice daily for one week. In nonketotic animals, mean blood glucose concentration decreased significantly during treatment with either DIPA or its acid moiety, sodium dichloroacetate. Neither di-isopropylammonium hydrochloride nor saline reduced hyperglycemia in the diabetic rats. There was no change in the blood glucose of nondiabetic rats in response to DIPA. In ketotic diabetes produced by streptozotocin, 63 per cent of DIPA-treated animals survived after one week, compared to 24 per cent of those given saline. DIPA did not influence glucose absorption in isolated everted jejunal sacs from diabetic and normal rats, while phenformin inhibited transport by more than 50 per cent Thus DIPA, which only partially mimics insulin in its actions and does not stimulate insulin release as do the sulfonylureas, also fails to duplicate the effects of phenformin and appears to have a unique mode of action. DIPA is an effective oral hypoglycemic agent in streptozotocin-diabetic rats.

Longitudinal biochemical and histologic studies were carried out in 11 children receiving oral hypoglycemic agents. There were five “suspected” diabetics (with evidence of glucose intolerance but without repeated fasting hyperglycemia) and six diabetics. Mean fasting plasma glucose (FPG) values showed no significant change during treatment with phenformin alone. The mean FPG decreased significantly within six to 10 weeks after addition of tolazamide to the regimen, but the decrease was not sustained during long-term observation (seven months to four years). Glucose disappearance rate (K) generally increased as FPG decreased, but the number of observations was smaller, and mean values showed no significant change. Mean values for fasting plasma insulin and for peak insulin response to intravenously administered glucose did not change significantly. Changes in capillary basement-membrane thickness (BMT) were found to be statistically significant in a number of individual instances. Decreasing BMTs were associated with increasing Ks and vice versa. A similar trend was apparent among other patients, in whom individual changes in BMT were not statistically significant. The pooled data were therefore subjected to chi-square analysis; K and average BMT were found to change in opposite directions (P < 0.001); a similar relationship held for K and minimum BMT (P < 0.05). But K and BMT did not correlate well as regards magnitude of change. Influences of phenformin and tolazamide on these changes could not be evaluated. Both BMT and K may be influenced by many complex variables, but the present findings indicate that glucose tolerance and BMT have a close interdependence.

Recent studies have demonstrated that phenformin inhibits intestinal glucose absorption in man. In order to determine if carbohydrate malabsorption occurs during phenformin administration we used a breath hydrogen (H2) test. This test depends upon fermentation of unabsorbed carbohydrate to hydrogen by colonic bacteria with subsequent absorption of H2 and pulmonary excretion. H2 excretion rates were measured in five normal subjects while fasting and two hours after the administration of 75 gm. of glucose. The study was repeated on the following day after the administration of 100 mg. of phenformin. There was no significant change in H2 excretion. One subject was then studied several times a day while eating a normal diet; phenformin administration did not increase H2 excretion. We conclude that the drug does not induce carbohydrate malabsorption.

Using uniformly labeled 14C L-lactate, we have studied the turnover and oxidation of lactic acid in a patient who presented with a mild lactic acidosis while on phenformin medication. As with other cases of lactic acidosis associated with phenformin therapy, this subject had impaired renal function as evidenced by serum creatinine levels of 2 mg./100 ml. and BUNs of 40 mg./l00 ml. Comparison of the rate of lactate oxidation relative to the rate of lactate turnover in this subject while on and off phenformin therapy suggests that a prime factor leading to the elevated lactate levels in this situation is impaired peripheral aerobic metabolism. Although lactate oxidation was increased in the presence of phenformin, the control studies clearly demonstrate that aerobic metabolism was not keeping pace with the increased level of anaerobic carbohydrate metabolism brought on by the drug. It is concluded that it is this imbalance in lactate metabolism that is responsible for the lactic acidosis that accompanies phenformin therapy.

Phenformin concentrations were measured in serum from seven patients with phenformin-associated lactic acidosis, and initial values ranging from 20 to 625 ng./ml. were obtained. Five of the seven patients had serum concentrations within the usual therapeutic range of up to 241 ng./ml. Serum phenformin concentrations were measured serially, and apparent half-lives of 5, 25, and 30 hours were obtained in three patients with serum creati nine concentrations of 1.7, 7.6, and 6.0 mg./dl., respectively. Although the half-life of phenformin was prolonged in azotemic patients, no correlation between serum creatinine concentration and serum phenformin could be demonstrated; furthermore, the severity of lactic acidosis as measured by arterial pH and lactate concentration did not correlate with the serum creatinine concentration.

Nineteen maturity-onset diabetic patients receiving oral hypoglycemic therapy in a university diabetes dinic completed a study to assess the efficacy of the oral agents and to determine their effects on pancreatic islet hormone secretion. All patients were receiving sulfonylureas, and seven were also receiving phenformin. Hie subjects were studied as outpatients in the clinic setting on four different occasions with collections of a baseline blood sample before a standard breakfast and a second sampling two hours postprandially, twice while on their prescribed medication and twice after having been withdrawn from the medication. The values obtained during the two studies on and the two studies off medications were reproducible for each subject. Analysis of the results by paired differences revealed that mean 24-hour urine glucose values deteriorated significantly (p < 0.005) after oral antidiabetic therapy was withdrawn; similarly, mean plasma glucose values, both at baseline and two hours postprandially, rose significantly (p < 0.001) when subjects were off medication. Baseline serum insulin values were not changed, but postprandial levels were significantly higher on oral agents (p It is apparent from this study that oral hypoglycemic medications can play a role in controlling symptoms in maturity-onset diabetic patients and that the beneficial effect of these agents on hyperglycemia may, in part, be explained by their stimulation of endogenous insulin secretion and partial suppression of endogenous glucagon.

We report 18 consecutive phenformin-treated diabetic patients admitted to this Medical Service acutely ill with metabolic acidosis. Lactic acidosis was anticipated, and documented, in all. Also, however, though most of the patients had only weakly positive, or even negative, serum reactions with the nitroprusside reagent, all were found to have coexisting ketoacidosis, plasma 3-hydroxybutyrate averaging 7.1 mmol/L. ± 3.9 (S.D.). This finding suggest that treatment of these patients should include insulin, and often also glucose, because most do not have marked hyperglycemia and some have hypoglycemia. The lactic acidosis in the nine patients who survivied was, on average, less severe than in the nine who died, but the difference was not statistically significant. Surivival correlated closely with the absence of shock on arrival. Only eight patients had a identifiable acute illness other than the metabolic acidosis. The other 10 patients had no discernible cause for the acute illness apart from their treatment with phenoformin. This finding raises serious doubts about whether phenformin should be used to treat patients with diabetes.

A patient with phenformin-associated lactic acidosis was treated with insulin and showed marked improvement coincident with the expected onset of action of the insulin administered. Relative insulin deficiency was demonstrated although several phenomena characteristic of phenformin-associated lactic acidosis obscured its reflection in the usual indices. From data presented and a review of the literature the following pathogenesis is proposed for the observed metabolic derangement. A background of relative insulin deficiency would permit enhanced pyruvate (and hence lactate) formation from protein sources. Insulin deficiency would also lead to inhibition of pyruvate dehydrogenase which slows pyruvate removal. Phenformin accumulation (cf impaired renal function) further reduces pyruvate removal by decreasing its conversion to glucose, but in addition alters the redox state. For the lactic acidosis which results, insulin administration may thus constitute specific therapy. Diabetes 24:28-35, January, 1975.

Phenformin administered intraperitoneally in doses of 30 and 100 mg. per kilogram significantly reduced the blood glucose concentration in streptozotocin-diabetic rats, but was virtually inactive in normal healthy rats. Phenformin also antagonized the hyperglycemic effect of an intravenous infusion of epinephrine in normal anesthetized rats and decreased blood glucose concentration of hepatectomized rats infused with glucose. These results are not consistent with hypotheses explaining the biguanide effect as an impairment of the intestinal absorption of sugars and indicate that the liver is not the only site of its action.

Studies of gluconeogenesis were carried out with isolated tubules prepared from the renal cortex of fasted rats. A study of the metabolic characteristics of this preparation upon incubation at 37° revealed a sustained respiratory rate, a constancy of ATP levels, and a linear rate of gluconeogenesis. Virtually no glucose synthesis was observed unless appropriate substrates were added; succinate and α-ketoglutarate were the most effective precursors. The addition of phenformin to the suspension of isolated tubules led to a decline of O2 uptake and ATP content in the absence and in the presence of added substrate. In the presence of succinate, there was a close correlation between the ATP content and the rate of gluconeogenesis. The data indicate that the phenformin-induced inhibition of gluconeogenesis is secondary to limitation of ATP generation rather than a direct effect on the enzymatic sequence leading to new glucose formation or to limited availability of reducing equivalents.

The purpose of the experiments described in this paper was to determine whether phenformin altered the metabolism of glucose in subjects in whom it did not lower the blood sugar. Glucose-1-C-14 was utilized to investigate the metabolism of glucose in eight normal and eight obese subjects before and after phenformin. Phenformin increased the turnover, recycling and conversion of glucose to lactate and decreased the oxidation of glucose in all of the normal subjects studied. The increase in glucose turnover was due entirely to the increase in glucose recycling since net glucose turnover (glucose derived from all sources except the Cori cycle) did not change. Phenformin also increased glucose turnover and recycling in obese subjects but in contrast to normals, its effect.s on glucose oxidation and glucose conversion to lactate were more variable. These results establish that phenformin has clearly defined effects on the metabolism of glucose in subjects in whom it exerts no hypoglycemic action and support the concept that phenformin increases glucose utilization by diverting it from oxidative to nonoxidative pathways in peripheral tissues.

1. Several methods of quantifying muscle capillary basement membrane width have been evaluated with the aim of determining whether variations in technic may influence the sensitivity of detecting significant capillary basement membrane thickening in diabetic subjects. 2. The results demonstrate that fixation of tissue in osmic acid followed by determination of mean basement membrane thickness reveals significant basement membrane thickening in over 90 per cent of diabetic subjects. 3. The error in estimation of mean basement membrane width due to oblique sectioning of diabetic capillaries is shown to be small, i.e. approximately 10 per cent. By contrast, measuring only the minimum basement membrane thickness of a capillary may underestimate diabetic basement membrane hypertrophy by from 20 to 40 per cent. 4. As compared to osmic acid fixation, glutaraldehyde leads to a very significant increase in the basement membrane width affecting normal capillaries relatively more than diabetic capillaries. 5. As a result, either fixation in glutaraldehyde or measurement of minimum basement membrane thickness may lead to a marked decrease in the sensitivity of the quantitative basement membrane technic. A procedure that employs both glutaraldehyde fixation and measurement of only minimum basement membrane width would appear to detect microangiopathy in only 45 per cent of diabetic subjects.

(a) From 67 to 87 per cent of the abnormal triamcinolone glucose tolerance curves are improved after treatment with small doses of hypoglycemic drugs. (b) As the weight reducing diet normalized only 6.6 per cent of the abnormal triamcinolone curves, it is possible to conclude that weight reduction alone is not enough to normalize this test. (c) The incidence of obstetrical complications (macrosomias, miscarriages, stillbirths, habitual abortion, and congenital malformations) in women with an abnormal triamcinolone test decreases remarkably or disappears after normalization of this test.

The effect of phenformin on the induction of plasma triglyceride by high-carbohydrate feeding was studied in ten subjects (nine normals and one Type II, hyperlipoproteinemia). Each study was divided into four periods. During periods I and III, each subject received an American Diabetes Association diet calculated to maintain body weight and in periods II and IV a high-carbohydrate diet (85 per cent of the calories derived from carbohydrate) which was isocaloric with the control diet. Phenformin, 50 mg. twice daily, was administered in either period II or IV in association with the high-carbohydrate diet. There was no change in weight during the study. The plasma triglyceride increase (ΔTG) during the high-carbohydrate diet alone was 103 mg. per cent as compared with a ΔTG of 26 mg. per cent during the high carbohydrate-phenformin period, a highly significant difference (p < 0.005). Insulin secretion and glucose tolerance during each period was determined by the sum of the increments above base baseline of plasma insulin and glucose concentrations during a five-hour oral glucose tolerance test. No significant differences in glucose tolerance or in insulin secretion occurred during any of the periods despite the alterations produced by the high-carbohydrate diet and phenformin in plasma triglyceride concentrations. A slight correlation could be demonstrated between the insulin response to a glucose load while on the control diet and the magnitude of the triglyceride response on the high-carbohydrate diet. The data indicate that (1) phenformin inhibits the induction of plasma triglyceride produced by carbohydrate loading, (2) there was no correlation between the changes in plasma triglyceride concentrations, glucose tolerance or insulin responsiveness during the experimental periods and (3) the magnitude of the increase in plasma triglyceride concentration may be related to the insulin response to a glucose load during the control period.

Oral and intravenous glucose tolerance tests were performed in ten normal subjects. Phenformin pretreatment flattened the glucose and insulin response to oral glucose loading but did not significantly alter the response to intravenous glucose. It is suggested that these results can best be explained by an inhibitory effect of the drug on the rate of intestinal glucose absorption.

A questionnaire was sent to 2,000 members of the American Diabetes Association during June-September, 1962. It concerned the aims and methods of present day therapy of diabetes mellitus, with special emphasis on therapy with oral agents. The answers were analyzed with respect to size of practice as well as other features. A total of 405 replies were received, comprising about 20,000 patients treated per year. Six years after the introduction of oral drugs into the United States, 30 per cent of diabetic patients are being treated with such agents. The aims of therapy are not altered from those for insulin treated patients. There is a noticeable lag in recourse to therapy with a combination of phenformin and a sulfonylurea drug after failure with a single drug. The advantage of use of such combinations in preserving the efficacy of oral therapy is pointed out. No significant incidence of toxicity or effect upon diabetic complications of these drugs alone or in combination is reported.