Observational follow-up of the PROactive study: a 6-year update.

AIMS: The PROactive study investigated pioglitazone for secondary prevention of macrovascular events in type 2 diabetes mellitus. Pioglitazone showed a 10% (non-significant) relative risk (RR) reduction for the primary composite endpoint and a significant 16% reduction for the main secondary endpoint (death, myocardial infarction, stroke) after a mean 34.5 months. There was no difference in cumulative malignancy incidence, but an imbalance in bladder malignancies (pioglitazone 14, placebo 5). We present a pre-specified 6-year interim analysis of a 10-year observational follow-up. METHODS: Any patient completing PROactive was eligible. No study treatments were provided. A Cox proportional hazard model compared non-adjudicated macrovascular events (same endpoints as PROactive excluding acute coronary syndrome) based on original randomization. Malignancies were compared using conventional RR ratios. RESULTS: Of 5238 randomized patients, 3599 (74%) entered the follow-up. For the follow-up (mean 5.8 years) or combined double-blind and follow-up periods (≤9.5 years, mean 8.7), there were no statistically significant differences in primary or main secondary endpoints. For the combined period, a similar percentage of patients had any diagnosed malignancy (RR = 1.05, 95% CI [0.89, 1.24]) or bladder malignancy (RR = 1.06, 95% CI [0.59, 1.89]) in the pioglitazone and placebo groups. There were fewer cases of bladder malignancy with pioglitazone (15 [0.6%] vs. 19 [0.7%] for placebo) for the combined period when events diagnosed in the first 365 days were excluded, and fewer cases for the follow-up period alone (10 [0.5%] vs. 17 [1.0%] for placebo). Further analyses of pioglitazone use (including use during follow-up) found no significant difference in bladder malignancies between any and no pioglitazone use for the combined period. CONCLUSIONS: These data suggest that improved macrovascular outcomes seen with pioglitazone subside without continued pioglitazone treatment. The double-blind period bladder cancer imbalance did not persist in follow-up.

High-density lipoprotein-cholesterol and not HbA1c was directly related to cardiovascular outcome in PROactive.

AIM: In PROactive, pioglitazone reduced the incidence of death, myocardial infarction and stroke, and significantly improved HbA1c, systolic blood pressure (SBP), triglycerides and high-density lipoprotein (HDL)-cholesterol relative to placebo. As these glycaemic and lipid parameters are major cardiovascular (CV) risk factors, we assessed their separate contribution to the reduced incidence of CV outcomes. METHODS: Patients (n = 5238) with type 2 diabetes and macrovascular disease were randomized to 45 mg pioglitazone or placebo. Relationships among treatment, outcome (time to first event of all-cause mortality, myocardial infarction and stroke) and 10 laboratory measurements and vital signs were investigated using log-linear models. Continuous variable measurements (percent changes from baseline to average of all postbaseline values prior to censoring) were made discrete by categorizing into tertiles. Log-linear models were fitted to multiway tables of discrete data and analysis of deviance used to summarize sources of variation in the data. RESULTS: Although pioglitazone treatment was associated with a decrease in HbA1c and an increase in HDL-cholesterol (HDL-C), only the change from baseline HDL-C predicted the outcome (χ(2) = 28.89, p < 0.0001). No other variables, including HbA1c, triglycerides and systolic blood pressure, showed significant direct associations with outcome. When the analysis was extended to include baseline statin use, this was associated with an improved outcome independently of HDL-C changes. CONCLUSIONS: This post hoc analysis suggests that HDL-C, but probably not HbA1c, is a driver of pioglitazone's favourable influence on CV outcome.

Specifically decreased collagen biosynthesis in scurvy dissociated from an effect on proline hydroxylation and correlated with body weight loss. In vitro studies in guinea pig calvarial bones.

The question whether ascorbate regulates collagen production solely through its direct role in proline hydroxylation was investigated. Proteins in calvarial bones from control and scorbutic weanling guinea pigs were labeled in short-term cultures with radioactive proline. Proteins were digested with purified bacterial collagenase to distinguish between effects on collagen polypeptide production and hydroxyproline formation. There was a preferential decrease in the absolute rate of collagen biosynthesis beginning after 2 wk of ascorbate deficiency, and this effect was temporally dissociated from decreased proline hydroxylation. There were no significant changes in the absolute rates of collagen degradation or noncollagen protein production. In vitro inhibition of proline hydroxylation in normal bone with alpha, alpha'-dipyridyl did not affect the relative rate of collagen synthesis, further dissociating these functions. Ascorbate added to scorbutic bone cultures reversed defective proline hydroxylation but not defective collagen synthesis, suggesting that the latter was an indirect effect of scurvy. There was a linear correlation between the extent of body weight lost during the 3rd and 4th wk of scurvy and the rate of collagen synthesis in scorbutic bone. This correlation also applied to control animals receiving ascorbate, but with weight loss induced by food restriction. These studies establish for the first time that ascorbate deficiency in guinea pigs leads to a specific decrease in collagen polypeptide synthesis and suggest that this decrease results from the reduced food intake and/or weight-loss characteristic of scurvy.

Decreased collagen production in diabetic rats.

Many of the chronic complications of diabetes mellitus involve defects in the connective tissue such as poor wound healing, diminished bone formation, and decreased linear growth. Because collagen is the major protein component of these connective tissues, we examined collagen production in diabetic rats as a probe of this generalized defect in connective tissue metabolism. Doses of streptozocin ranging from 35 to 300 mg/kg were used to induce diabetes of graded metabolic severity in rats. Parietal bone or articular cartilage was removed and incubated at 37 degrees C with 5 microCi L-[5-3H]proline for 2 h, and collagen and noncollagen protein production were quantitated after separation with purified bacterial collagenase. Within 2 wk after induction of diabetes, collagen production was significantly reduced in bone and cartilage from diabetic rats to 52% (P less than .01) and 51% (P less than .01) of control (buffer-injected) levels, respectively. In contrast, noncollagen protein production in bone and cartilage from diabetic animals was no different from in tissue from control rats. The correlation between collagen relative to total protein production (relative rate) and the degree of hyperglycemia was highly significant for both bone (r = -.77, P less than .001) and cartilage (r = -.87, P less than .001). Other factors found to correlate with altered collagen production were the duration of diabetes and the amount of weight loss. Thus, diabetes is associated with a marked decrease in collagen production, which was seen early after induction of diabetes and was specific when compared with noncollagen protein production. Cumulative effects of these marked changes in collagen production may contribute to the chronic connective tissue complications in diabetes.

Nutritional and hormonal regulation of articular collagen production in diabetic animals.

Although changes in collagen production probably play a major role in the connective tissue defects of diabetes, we do not know to what extent these changes are attributable to hormonal/metabolic versus nutritional alterations. To study collagen production as influenced separately by nutrition versus hormonal/metabolic factors, rats were given 50 mg/kg i.v. streptozocin (STZ) (mild weight-gaining diabetes) or 100 mg/kg STZ (severe weight-losing diabetes) and compared with nondiabetic food-restricted rats to match weight changes in diabetic animals. Articular cartilage was incubated with [3H]proline, and uptake of [3H]proline into both collagen and noncollagen proteins was determined with purified bacterial collagenase. Collagen decreased to 49% in mildly diabetic rats and 16% in severely diabetic rats, compared with control rats fed ad libitum and decreased to 85 and 73%, respectively, in food-restricted rats (both P less than .01 vs. diabetes). Diabetes induced a greater defect in collagen production than food restriction and a greater decrease in collagen than noncollagen protein production within each group, suggesting a specific effect on collagen. With comparable levels of metabolic severity (glucose, beta-hydroxybutyrate), diabetic animals that lost weight produced significantly less collagen than animals that gained weight, suggesting separate mechanisms. Quantitation of the impact of undernutrition on collagen production in diabetes demonstrated that approximately 31 to 32% of the defect was due to undernutrition, leaving approximately 68-69% of the defect due to the diabetic state.(ABSTRACT TRUNCATED AT 250 WORDS)

Tissue-specific regulation of basic fibroblast growth factor mRNA levels by diabetes.

Because basic fibroblast growth factor (bFGF) is recognized as an angiogenic factor and diabetes is characterized by multiple vascular complications, including diabetic microangiopathy, we examined the regulation of tissue bFGF mRNA levels by diabetes. Diabetes was induced in male Sprague-Dawley rats by injection of 125 mg/kg body wt i.v. streptozocin (STZ), with intensive insulin therapy initiated in half of the diabetic rats. Rats were killed 96 h postinjection of STZ. Tissue bFGF and insulinlike growth factor I (IGF-I) mRNA levels were measured simultaneously with a solution hybridization-RNase protection assay. bFGF mRNA levels increased from 1.7- to 2.7-fold in eye, heart, lung, and brain from diabetic compared with buffer-injected control rats. In skeletal muscle, bFGF mRNA levels decreased to 23% of control levels, whereas bFGF mRNA levels were unchanged in kidneys from diabetic versus control rats. Changes in tissue bFGF mRNA levels were partially reversed by insulin treatment in all tissues. In contrast, IGF-I mRNA levels were significantly decreased from 15 to 50% of control levels in all tissues studied except those in brain, which decreased to only 85% of control levels. These data demonstrate that bFGF mRNA levels are altered by diabetes in a tissue-specific fashion and are consistent with the hypothesis that increased production of bFGF may contribute to the development of diabetic microangiopathy in some tissues.

Differential binding of insulin to human arterial and venous endothelial cells in primary culture.

The properties of 125I-insulin binding were assessed in endothelial cells prepared from the veins and the arteries of human umbilical cords. The endothelial nature of both the natural and venous cultures were documented by the presence of characteristic endothelial features, including Weibel-Palade bodies, factor VIII antigen, and morphology. Both arterial and venous cells possessed typical receptors for insulin on the basis of specificity of binding, curvilinear Scatchard plots, affinity profiles, pH dependency, and dissociation kinetics. Arterial cells bound at least 2.5 times more insulin than did venous cells, whether studied at 4 h, 24 h, or 72 h after in vitro plating. We conclude that (1) specific receptors for insulin are present on human arterial as well as human venous endothelial cells and (2) the concentration of insulin receptors varies among endothelial cells derived from different vascular sources.

Hyperthyroidism caused by inappropriate thyrotropin hypersecretion: studies in patients with selective pituitary resistance to thyroid hormone.

We have identified the condition of thyrotropin (thyroid-stimulating hormone [TSH])-induced hyperthyroidism secondary to selective pituitary insensitivity to thyroid hormone in three patients. Each patient was clinically hyperthyroid, with elevated serum levels of thyroxine (T4) and triiodothyronine (T3) and detectable levels of serum TSH before therapy. After therapy each patient had notably elevated TSH levels at a time that peripheral levels of thyroid hormones were in the hyperthyroid range. Before and after therapy, serum levels of TSH were suppressed by therapy with liothyronine sodium and were stimulated by protirelin (thyrotropin-releasing hormone) both before and after liothyronine and dexamethasone treatment. Dexamethasone therapy decreased the levels of TSH, protirelin-stimulated TSH, and circulating T4 and T3. Serum levels of glycoprotein alpha-subunit were 0.6 to 2.4 ng/ml, values considerably lower than found in patients with TSH-secreting pituitary tumors. We suggest that the frequency of TSH-induced hyperthyroidism secondary to pituitary insensitivity to thyroid hormone may be higher than presently indicated in the medical literature.

Prediction of glucose response to weight loss in patients with non-insulin-dependent diabetes mellitus.

Although diet therapy is considered the cornerstone of therapy for obese patients with non-insulin-dependent diabetes mellitus, losing weight is often difficult, and the plasma glucose concentration does not always improve after weight loss. We looked for predictors of improvement in plasma glucose levels after weight loss in 135 obese patients with non-insulin-dependent diabetes mellitus who had lost at least 9.1 kg of body weight. After weight loss there was a bimodal distribution of plasma glucose levels, allowing us to identify patients as "responders" or "nonresponders" according to whether a random plasma glucose level was above or below 10.0 mmol/L after a 9.1-kg weight loss. Fifty-five (41%) of 135 patients were responders (after a 9.1-kg weight loss, the mean +/- SEM plasma glucose level was 7.0 +/- 0.2 mmol/L). Many responders had improved plasma glucose levels after only slight weight loss. Eighty (59%) of 135 patients were nonresponders (after a 9.1-kg weight loss, the mean +/- SEM plasma glucose level was 18.3 +/- 0.6 mmol/L). Although the responder and nonresponder groups were comparable in age, sex distribution, plasma glucose levels, and body weight at initial presentation, improvement in the plasma glucose level after weight loss could be predicted by a plasma glucose level of 10.0 mmol/L or lower after 2.3-kg (62% positive predictive value) and 4.5-kg (79% positive predictive value) weight loss. We conclude that, in contrast to conventional teaching, many patients with non-insulin-dependent diabetes mellitus will not have any improvement in plasma glucose levels after a 9.1-kg weight loss. However, a substantial minority (approximately 40%) of obese patients with non-insulin-dependent diabetes mellitus have much lower plasma glucose levels with a weight loss of 9.1 kg or less. Although the plasma glucose response to weight loss cannot be forecast by initial clinical parameters, the success or failure of diet therapy can be predicted from the plasma glucose level after a weight loss of only 2.3 to 4.5 kg. Mild or moderately obese patients with non-insulin-dependent diabetes mellitus who remain hyperglycemic after a weight loss of 2.3 to 9.1 kg are unlikely to improve with further weight loss and should be considered for treatment with insulin or oral hypoglycemic agents.

Inhibition of collagen production by diabetic rat serum: response to insulin and insulin-like growth factor-I added in vitro.

Diabetes mellitus is associated with a generalized defect in connective tissue metabolism. Since collagen is the major protein of connective tissues, we used collagen as a probe to examine the role of factors in diabetic rat serum (DRS) in the etiology of these defects. Serum and skin fibroblasts were isolated from nondiabetic rats, and serum was taken from rats 48 h after injection of 200 mg/kg streptozotocin. Within 24 h of confluency, the fibroblast medium was changed to experimental serum for 24 h, with 5 microCi [3H]proline added for the final 6 h. Collagen and noncollagen proteins were quantitated using purified collagenase. Compared to cells incubated in medium without serum, collagen fell to 58% with 0.5% DRS (P less than 0.05) and continued to decrease with increasing concentrations of DRS. Noncollagen protein decreased below levels in cells incubated in medium without serum only when concentrations of diabetic serum were 1% or greater and did not decrease further with higher concentrations of diabetic serum. Collagen was decreased to a greater degree than noncollagen protein at each concentration of DRS, such that collagen relative to total protein production was significantly reduced at 0.5% or more DRS. Addition of 10(-7)-10(-9) M insulin or insulin-like growth factor-I (0.1-1000 ng/ml) to DRS did not return collagen production to the level seen in cells incubated in medium with no added serum (basal production). After separation of serum components based on size, incubation of cells with the low mol wt fraction (less than 5000 daltons) of normal and diabetic rat serum resulted in equivalent collagen production, while incubation with the high mol wt fraction of DRS resulted in 200-fold less collagen compared to the similar fraction of normal serum. This decrease in collagen production appeared due to the presence of a high mol wt factor(s) in diabetic serum which had a direct inhibitory effect on collagen and was not due to deficiency of growth peptides. The degree and specificity of these changes in collagen production probably contribute to long term complications in diabetes through altered connective tissue metabolism.

Multiplication-stimulating activity (MSA) stimulates proteoglycan synthesis in cultured endothelial cells.

Cultured endothelial cells have previously been shown to have specific binding sites for certain insulin-like growth factors (IGFs) and insulin. In the present study, the IGF, Multiplication Stimulating Activity (MSA), at concentrations of 100-1000 ng/ml, stimulated 35SO4 incorporation into proteoglycans by endothelial cells cultured from bovine pulmonary arteries. The stimulatory effect on proteoglycan synthesis was not accompanied by changes in cell number, total protein synthesis, nor collagen synthesis and was not mimicked by insulin. These findings suggest that MSA specifically stimulated accumulation of proteoglycans in the cultured endothelial cells, with the effect mediated through an IGF receptor.

Comparison of insulin binding to cells of fed and fasted obese patients: results in erythrocytes and monocytes.

The erythrocyte (RBC) has received recent interest as a cell model to examine insulin receptor status in humans. In the present study we have compared the insulin receptors on mature RBCs and monocytes of four hyperinsulinemic obese patients in the fed state and after a 14-day fast (less than 50 cal/day). Insulin binding in the basal (fed) state was described in RBC and monocytes due predominantly to a decrease in the receptor concentration in both cell types. After a 14-day fast, insulin binding to both RBCs and monocytes increased significantly in each patient. Maximal binding of [125I]iodoinsulin to RBCs increased by 29% (range, 20-46%), and binding to monocytes increased by 116% (range, 46-321%). In response to the fast, the concentration of insulin needed to inhibit binding by 50% decreased from 5 to 2 ng/ml in RBC and from 3 to 1 ng/ml in monocytes. Conventional and computer-fitted Scatchard analyses demonstrated no change in the receptor concentration of RBCs of any patient, whereas the receptor concentration of monocytes increased by more than 50% in two of the four patients and by 40% for the group. Thus, in response to the fast, the direction of the change in insulin binding was similar in the RBCs and monocytes, whereas the magnitude and, in certain patients, the mechanism of the binding increase differed.

Sulfated glycosaminoglycans in cultured endothelial cells from capillaries and large vessels of human and bovine origin.

The [35S]glycosaminoglycans ([35S]GAG) synthesized by capillary endothelial cells were analyzed and compared to GAG synthesized by endothelial cells cultured from 4 larger vessels. Two separate cultures of endothelial cells were established from bovine fat capillaries and from 4 larger vessels of human origin (umbilical vein) and bovine origin (pulmonary artery, pulmonary vein and aorta). After incubation with 35SO4 for 72 h, the [35S]glycosaminoglycans (GAG) composition of the media, pericellular and cellular fractions of each culture were determined by selective degradation with nitrous acid, chondroitinase ABC and chondroitinase AC. All endothelial cells produced large amounts of [35S]GAG with increased proportions of heparinoids (heparan sulfate and heparin) in the cellular and pericellular fractions. Each culture showed a distinct distribution of [35S]GAG in the media, pericellular and cellular fractions with several specific differences found among the 5 cultures. The differences in GAG content were confirmed in a second group of separate cultures from each of the 5 vessels indicating that, although having several features of GAG metabolism in common, each endothelial cell culture demonstrated a characteristic complement of synthesized, secreted and cell surface-sulfated glycosaminoglycans.

Direct inhibition of collagen production in vitro by diabetic rat serum.

Diabetes mellitus is associated with a generalized defect in connective tissue metabolism, including decreased growth, poor wound healing, and osteopenia. To determine the role of circulating factors in the etiology of these defects, we studied the effects of diabetic rat serum (DRS) on collagen, the major protein component of connective tissues. After preincubation of costal cartilage from hypophysectomized rats with experimental serum for 20 hours, [3H] proline was added for final four hours of incubation. Collagen and noncollagen protein were quantitated using purified bacterial collagenase. Compared to incubation of tissue in buffer without added serum, collagen production in cartilage incubated with 2% DRS was decreased by 23% (P less than .05), and with 4% serum by 88% (P less than .01). In contrast, serum from normal rats (NRS) increased collagen to 158% above buffer-incubated cartilage at 1.0% (P less than .02) and to 196% at 2% serum (P less than .01). Noncollagen protein production decreased below buffer only after addition of 2% or more DRS and increased above buffer after addition of 2% or more of NRS (178%, P less than .05). Addition of insulin at 10 and 100 mU/mL to DRS did not reverse defective collagen production, and addition of glucose (up to 900 mg/dL) or ketones (20 mmol/L) to NRS did not induce the changes in collagen production seen after addition of diabetic serum. Chromatographic separation of serum revealed that the inhibitory activity of DRS was in the high molecular weight fraction (less than 5000).(ABSTRACT TRUNCATED AT 250 WORDS)

Regulation of growth factor mRNA levels in the eyes of diabetic rats.

The underlying etiology of diabetic microvascular disease remains unknown. To examine the potential contribution of basic fibroblast growth factor (bFGF), which is an angiogenic factor, and insulin-like growth factor-I (IGF-I) to the development of diabetic microvascular disease, bFGF and IGF-I mRNA levels were measured in tissues of control, diabetic, and insulin-treated diabetic rats. Diabetes was induced in rats by intravenous injection of streptozotocin (STZ) 65 mg/kg, and the rats were maintained for 21 days. bFGF mRNA levels increased threefold in the eyes of diabetic versus control rats, whereas a consistent change in bFGF mRNA levels was not observed in other tissues. In contrast, IGF-I mRNA levels decreased in the eyes and other tissues, including kidney, lung, and skeletal muscle, of diabetic as compared with control rats. Insulin treatment prevented the diabetes-induced increase in bFGF and decrease in IGF-I mRNA levels. Acidic FGF (aFGF) mRNA levels were unchanged in eyes from diabetic versus control rats. In partially purified retinas, diabetes increased bFGF mRNA levels twofold as compared with levels in control retinas, whereas IGF-I mRNA levels decreased to 58% of control levels in retinas from diabetic rats. Insulin treatment again prevented the diabetes-induced increase in IGF-I mRNA levels in the retina but had no effect on the diabetes-induced increase in bFGF mRNA levels. bFGF peptide levels were minimally increased in diabetic versus control retinas.(ABSTRACT TRUNCATED AT 250 WORDS)

Reducing cardiovascular risk in diabetes. Which factors to modify first?

Up to 80% of diabetic patients die of macrovascular complications, including CAD, stroke, and peripheral vascular disease. Because of the growing numbers of diabetic patients and the increased mortality after their first cardiovascular event, it is critical to identify and treat risk factors early and aggressively in these patients. Numerous studies in patients with type 2 diabetes have shown the benefits of aggressive treatment of blood pressure and lipids to levels that 10 years ago would have seemed abnormally low. The downward changes in "normal" limits can be frustrating to primary care physicians, but advances in treatment are redefining "normal" levels required to avoid complications in this high-risk population.

Differential regulation of Ota and Otb, two primary glycine betaine transporters in the methanogenic archaeon methanosarcina mazei Gö1.

Methanogenic archaea accumulate glycine betaine in response to hypersalinity, but the regulation of proteins involved, their mechanism of activation and regulation of the corresponding genes are largely unknown. Methanosarcina mazei differs from most other methanoarchaea in having two gene clusters both encoding a potential glycine betaine transporter, Ota and Otb. Western blot as well as quantitative real-time PCR revealed that Otb is not regulated by osmolarity. On the other hand, cellular levels of Ota increased with increasing salt concentrations. A maximum was reached at 300-500 mM NaCl. Ota concentrations reached a maximum 4 h after an osmotic upshock. Hyperosmolarity also caused an increase in cellular Ota concentrations. In addition to osmolarity Ota expression was regulated by the growth phase. Expression of Ota as well as transport of betaine was downregulated in the presence of glycine betaine.

Correlation between decreased collagen production in diabetic animals and in cells exposed to diabetic serum: response to insulin.

Collagen production has been shown to be decreased in costal cartilage from nondiabetic animals after incubation with diabetic rat serum. Since collagen was decreased to a similar degree in tissues from diabetic animals, we questioned whether altered collagen production in vivo could be related to altered production induced in vitro. Collagen and noncollagen protein production in articular cartilage from diabetic animals (production in vivo) was compared to protein production in dermal fibroblasts from non-diabetic rats exposed to serum from the same diabetic rats (production in vitro). Diabetes was induced by intravenous administration of 90 mg/kg of streptozotocin into male Sprague-Dawley rats. Cartilage was removed and incubated with [3H]-proline for 2 hours at 37 degrees C (in vivo), while fibroblasts were exposed to experimental serum from individual animals for 24 hours with addition of 5 microCi of [5-3H]-proline for the final 6 hours (in vitro). Collagen and noncollagen protein production were quantitated using purified bacterial collagenase. Collagen production in cartilage decreased to 46% (p less than .01) and noncollagen to 68% (p less than .05) of levels in control animals. Fibroblasts exposed to 2.5% diabetic serum decreased collagen and noncollagen protein production to levels of 30% (p less than .01) and 54% (p less than .05) of production in cells incubated in 2.5% normal rat serum. Correlation between defective collagen production in cartilage from individual rats and the effects of their own serum on collagen production in fibroblasts was significant (r = 0.84, p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

A specific decrease in collagen synthesis in acutely fasted, vitamin C-supplemented, guinea pigs.

Weight loss often results from various experimental conditions including scurvy in guinea pigs, where we showed that decreased collagen synthesis was directly related to weight loss, rather than to defective proline hydroxylation (Chojkier, M., Spanheimer, R., and Peterkofsky, B. (1983) J. Clin. Invest. 72, 826-835). In the study described here, this effect was reproduced by acutely fasting normal guinea pigs receiving vitamin C, as determined by measuring collagen and non-collagen protein production after labeling tissues in vitro with [3H]proline. Collagen production (dpm/microgram of DNA) decreased soon after initiating fasting and by 96 h it had reached levels 8-12% of control values. Effects on non-collagen protein were much less severe, so that the percentage of collagen synthesis relative to total protein synthesis was 20-25% of control values after a 96-h fast. These effects were not due to changes in the specific radioactivity of free proline. Refeeding reversed the effects on non-collagen protein production within 24 h, but collagen production did not return to normal until 96 h. The effect of fasting on collagen production was independent of age, sex, ascorbate status, species of animal, and type of connective tissue and also was seen with in vivo labeling. Pulse-chase experiments and analysis of labeled and pre-existing proteins by gel electrophoresis showed no evidence of increased collagen degradation as a result of fasting. Procollagen mRNA was decreased in tissues of fasted animals as determined by cell-free translation and dot-blot hybridization with cDNA probes. In contrast, there was no decrease in translatable mRNAs for non-collagen proteins. These results suggest that loss of nutritional factors other than vitamin C lead to a rapid, specific decrease in collagen synthesis mainly through modulation of mRNA levels.

Phenytoin-induced agranulocytosis: a nonimmunologic idiosyncratic reaction?

Agranulocytosis is a rare side effect of phenytoin treatment. We describe the case of an elderly man who developed agranulocytosis 2 weeks following initiation of phenytoin treatment, on no cytotoxic drugs or any other medications except decadron. The white blood cell count was 300/mm3 with absent granulocytes. The liver enzymes were also noted to be newly elevated. Complete workup showed no evidence of infection. The patient was managed as a case of neutropenic fever and phenytoin was discontinued. A bone marrow biopsy showed absent granulocyte precursors after myelocytes, and an antigranulocyte antibody titer was negative, which suggests a direct toxic effect rather than an immunologic phenomenon. Both agranulocytosis and hepatitis were readily reversible on phenytoin cessation.