Moderate ingestion of alcohol is associated with acute ethanol-induced suppression of circulating CTX in a PTH-independent fashion

Abstract

The "J shape" curve linking the risk of poor bone health to alcohol intake is now well recognized from epidemiological studies. Ethanol and nonethanol components of alcoholic beverages could influence bone remodeling. However, in the absence of a solid underlying mechanism, the positive association between moderate alcoholic intake and BMD remains questionable because of confounding associated social factors. The objective of this work was to characterize the short-term effects of moderate alcohol consumption on circulating bone markers, especially those involved in bone resorption. Two sequential blood-sampling studies were undertaken in fasted healthy volunteers (age, 20-47 yr) over a 6-h period using beer of different alcohol levels (<0.05-4.6%), solutions of ethanol or orthosilicic acid (two major components of beer), and water +/- calcium chloride (positive and negative controls, respectively). Study 1 (24 subjects) assessed the effects of the different solutions, whereas study 2 (26 subjects) focused on ethanol/beer dose. Using all data in a "mixed effect model," we identified the contributions of the individual components of beer, namely ethanol, energy, low-dose calcium, and high-dose orthosilicic acid, on acute bone resorption. Markers of bone formation were unchanged throughout the study for all solutions investigated. In contrast, the bone resorption marker, serum carboxy terminal telopeptide of type I collagen (CTX), was significantly reduced after ingestion of a 0.6 liters of ethanol solution (>2% ethanol; p <or= 0.01, RM-ANOVA), 0.6 liters of beer (<0.05-4.6% ethanol; p < 0.02), or a solution of calcium (180 mg calcium; p < 0.001), but only after calcium ingestion was the reduction in CTX preceded by a significant fall in serum PTH (p < 0.001). Orthosilicic acid had no acute effect. Similar reductions in CTX, from baseline, were measured in urine after ingestion of the test solutions; however, the biological variability in urine CTX was greater compared with serum CTX. Modeling indicated that the major, acute suppressive effects of moderate beer ingestion (0.6 liters) on CTX were caused by energy intake in the early phase (approximately 0-3 h) and a "nonenergy" ethanol component in the later phase (approximately 3 to >6 h). The early effect on bone resorption is well described after the intake of energy, mediated by glucagon-like peptide-2, but the late effect of moderate alcohol ingestion is novel, seems to be ethanol specific, and is mediated in a non-calcitonin- and a non-PTH-dependent fashion, thus providing a mechanism for the positive association between moderate alcohol ingestion and BMD.

FIG. 1

Serum CICP (A), bALP (B), and osteocalcin (C), as a percentage of basal values, after ingestion of 0.6 liters of the following: 4.6% (vol/vol) beer containing 27.6 g ethanol, 22.5 mg Si, and 52.4 mg Ca, deionized water containing 21.2 mg Si with and without 180 mg Ca, and deionized water with 180 mg Ca alone. Silicon was in the form of OSA and Ca as calcium chloride. Becauase of cost, sample analysis was restricted to CICP and osteocalcin after ingestion of Ca or Ca + Si. Because beer ingestion had no effect on the formation markers, the other components of beer, namely 4.6% ethanol and alcohol-free beer, were not studied. Data are means ± SD of four subjects (two men and two women), except for serum CICP after Ca and Ca + Si ingestions, which are a mean of eight subjects (four men and four women).

FIG. 2

Serum CTX, as a percentage of basal values, after ingestion of 0.6 liters of the following: deionized water with and without 180 mg Ca (A), 4.6% (vol/vol) beer containing 27.6 g ethanol, 22.5 mg Si, and 52.4 mg Ca, alcohol-free beer containing 23 mg Si or deionized water with 27.6 g ethanol (B), and deionized water containing 21.2 mg Si with and without 180 mg Ca (C). Silicon was in the form of OSA and Ca as calcium chloride. The fall in CTX was significant after ingestion of beer, ethanol solution, alcohol-free beer, and Ca with and without Si (all p < 0.001; RM-ANOVA) compared with water. Data are means ± SD of eight subjects (four men and four women).

FIG. 3

Serum iPTH, as a percentage of basal values, after ingestion of 0.6 liters of the following: deionized water with and without 180 mg Ca (A), 4.6% (vol/vol) beer containing 27.6 g ethanol, 22.5 mg Si, and 52.4 mg Ca, or deionized water with 27.6 g ethanol (B), and deionized water containing 21.2 mg Si with and without 180 mg Ca (C). Serum calcitonin after ingestion of 0.6 liters deionized water with and without 27.6 g ethanol is also shown (D). Silicon was in the form of OSA and Ca as calcium chloride. Ingestion of Ca with or without Si led to a significant decrease in iPTH (p < 0.001; RM-ANOVA) compared with water. Data are means ± SD of eight subjects (iPTH; four men and four women) or five subjects (calcitonin; three men and two women).

FIG. 4

Urine CTX, as a percentage of basal values, after ingestion of 0.6 liters of the following: 4.6% (vol/vol) beer containing 27.6 g ethanol, 22.5 mg Si, and 52.4 mg Ca, deionized water containing 180 mg Ca with and without 21.2 mg Si, and deionized water with 21.2 mg Si. Silicon was in the form of OSA and Ca as calcium chloride. The reduction in urine CTX was significant (paired Student's t-test) for beer (p < 0.0001 at 0–3 h and p = 0.001 at 3–6 h), calcium (p = 0.02 at 0–3 h and p = 0.001 at 3–6 h), and calcium + silicon (p = 0.009 at 0–3 h and p = 0.0001 at 3–6 h) ingestions. The reduction in CTX at 3–6 h after ingestion of OSA was also significant (p = 0.015) and is likely to be caused by the normal diurnal variation in urine CTX.(9) Data are means ± SD of nine subjects (five men and four women) for beer and silicon ingestions and eight subjects (four men and four women) for calcium and calcium + silicon ingestions.

FIG. 5

Plasma CTX (A and B) and plasma alcohol (C and D) levels, as a percentage of basal values, after ingestion of 0.6 liters of the following: 4.1% (vol/vol) and 2.05% (vol/vol) ethanol (A and C) and 4.1% (vol/vol) and 2.05% (vol/vol) beer (B and D). Blood alcohol levels were markedly elevated (p < 0.001) after ingestion of beers and solutions of ethanol (p < 0.001; RM-ANOVA). The fall in CTX was significant for 4.1% ethanol (p = 0.01; RM-ANOVA) and for 2.05% and 4.1% beer (p ≤ 0.02; RM-ANOVA). Data are means ± SD of 8 (4 men and 4 women; A and C) and 18 subjects (9 men and 9 women; B and D); 0.6 liters of 4.1% or 2.05% beer or ethanol is equivalent to 24.6 and 12.3 g alcohol/ethanol.

FIG. 6

Mixed-effect modeling of the contribution of the major components of beer on the change in serum CTX (as a percent of basal value) with time after simulated ingestion of 0.6 liters of 4.6% (vol/vol) beer (27.6 g alcohol). Results have been adjusted for sex.