|
|
||||||||
Original Research Communications |
1 From the Departments of Nutritional Sciences and Paediatrics, University of Toronto; The Research Institute, The Hospital for Sick Children, Toronto; and the Department of Agricultural, Food, and Nutritional Sciences, University of Alberta, Alberta, Canada.
2 Presented in part at the annual meeting of the Federation of American Societies for Experimental Biology, New Orleans, April 1997. 3 Supported by grant MT 10321 of the Medical Research Council of Canada. DCW was supported by a Medical Research Council of Canada Fellowship and a research scholarship from Queen's University of Belfast. Mead Johnson Canada provided the protein-free powder for the experimental diets. 4 Address reprint requests to PB Pencharz, Division of Gastroenterology and Nutrition, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G 1X8. E-mail: paul.pencharz{at}sickkids.on.ca.
| ABSTRACT |
|---|
|
|
|---|
Objective: Our objective was to determine the requirement for threonine in men by using the indicator amino acid oxidation method and to provide statistical estimates of the population mean and 95% CIs of the threonine requirement. We hypothesized that the current World Health Organization estimate of the threonine requirement, 7 mgkg-1d-1 (based on nitrogen balance studies), is too low.
Design: Six healthy men each received 6 different threonine intakes while consuming an energy-sufficient diet with 1.0 g L-amino acid mixturekg-1d-1. The effect of graded alterations in dietary threonine intake on phenylalanine flux and oxidation was studied by using L-[1-13C]phenylalanine as the indicator amino acid.
Results: The results of two-phase linear regression crossover analysis showed that the mean threonine requirement, based on indicator oxidation, was 19.0 mgkg-1d-1 with an upper safe intake of 26.2 mgkg-1d-1.
Conclusions: This is the first application of the indicator amino acid oxidation technique in humans to study the requirement for an indispensable amino acid with a complex degradative pathway. We found that the upper safe intake for 95% of the population is almost 4-fold higher than the current World Health Organization estimate.
Key Words: Threonine indicator amino acid oxidation amino acid requirement stable isotope phenylalanine men
| INTRODUCTION |
|---|
|
|
|---|
Threonine is a totally indispensable amino acid because it has no metabolic precursor (12). The upper end of the range for recommended intake, 7 mgkg-1d-1 (1, 2), is based on nitrogen balance studies performed in only 3 healthy men (13). By using direct amino acid oxidation, Zhao et al (14) estimated the requirement for threonine to be between 10 and 20 mgkg-1d-1.
For the direct oxidation method to be completely reliable, the tracer amino acid must have only a single degradative pathway, oxidation to carbon dioxide (4). However, threonine has 2 degradative pathways in mammals (15, 16), and therefore this condition is not met. In contrast, threonine requirements in humans could be measured by IAAO because an alternative amino acid is used as the indicator and the method is therefore independent of the partitioning of carbon atoms from threonine degradation among several possible pools (15, 16). We proposed that use of the IAAO method would allow us to improve upon the current estimate of the threonine requirement in men. Our objectives were to determine the requirement for threonine in men by using the IAAO method and to provide statistical estimates of the population mean and 95% CIs for the threonine requirement.
| SUBJECTS AND METHODS |
|---|
|
|
|---|
|
Experimental design
We used a previously described experimental approach of IAAO studies (911). Briefly, the total experimental period for each subject was 18 d, divided into two 9-d periods during which the subjects were free-living, except when the oxidation studies were being performed. These 2 periods were separated by
2 wk but <1 mo to allow for a break from the experimental diet. On days 3, 6, and 9 of each dietary period, each subject was assigned (in random order) an amount of the test amino acid. During each dietary period, subjects consumed a diet containing 14 mg phenylalaninekg-1d-1 to ensure adequate dietary phenylalanine. This amount meets the requirement of 95% of adult males, as previously determined by amino acid oxidation when tyrosine was present in relative excess (8). It has been shown that the indicator amino acid is most sensitive to the intake of the test amino acid when the former is consumed at its required amount (17). The amino acid composition of the diet was based on egg protein and contained generous amounts of both tyrosine and threonine (40 and 47 mgkg-1d-1, respectively) (18).
The estimate of threonine requirement is 7 mgkg-1d-1 by nitrogen balance (1) and is between 10 and 20 mgkg-1d-1 by direct oxidation (14). We predicted an estimated requirement close to 20 mgkg-1d-1, on the basis of the assumption that the estimate from direct oxidation was low as a result of undetected partition of some oxidized threonine into glycine (16). On days 3, 6, and 9 of each dietary period, the subjects received a test intake of dietary threonine at 1 of 7 amounts (5, 10, 15, 20, 25, 30, or 35 mgkg-1d-1), assigned in random order, and then measurements of phenylalanine flux and oxidation were performed. Each subject was therefore studied at 6 different amounts (36 studies in all); thus, we had sufficient test threonine intakes for estimation of the population mean and 95% CIs. The distribution of the number of studies at each test amount is shown in Table 2
. The requirement breakpoint estimate depends particularly on the slope of the line for the lowest intake amount. To better define the slope of this line, we assigned 6 studies each to the 2 lowest amounts, which also meant it was only possible to assign 4 studies to each of the 2 highest amounts.
|
Experimental diet
The experimental diet was specifically developed for use in studies of amino acid metabolism (18) and has been used in such studies previously (811). All amino nitrogen was supplied as a crystalline amino acid mixture that simulated the composition of intact egg protein and was consumed in the amount of 1 g proteinkg-1d-1. This mixture was needed to maintain a phenylalanine intake equal to the requirement, thereby ensuring maximum sensitivity of the method (20), and to ensure that other amino acids in the diet were kept constant. This mixture was also needed to control threonine intake so that dietary amounts ranging from deficient to excessive, based on recommendations (1) and direct oxidation studies (14), could be studied. The remainder of the diet was provided as nonprotein energy sources and it met the energy, vitamin, and mineral requirements of the subjects. The main source of energy in the diet was a flavored liquid-formula diet (Protein Free Powder, product no. 80056; Mead Johnson, Evansville, IN), with the remainder of the energy supplied by 2 types of protein-free cookies prepared in the research kitchen of The Hospital for Sick Children (18). Supplemental foods were not permitted during the study, except for clear tea, decaffeinated coffee, and water. The diet was formulated to provide 10% of energy as protein, 53% as carbohydrate, and 37% as fat. Energy intake was calculated as described above and was constant for each individual during the study period.
Diets were prepared and weighed (model PE2000 Mettler scale; Mettler, Nanikon, Switzerland) in the research kitchen and were then portioned into 4 isoenergetic, isonitrogenous meals that were consumed daily at 0800, 1200, 1600, and 2000. Each meal consisted of a bottle of flavored, protein-free formula, into which the amino acid mixture had been added, and one of each type of protein-free cookie.
Anthropometry
Standing heights were measured to the nearest 0.1 cm by using a wall-mounted stadiometer (Holtain Ltd, Croswell, Crymych, United Kingdom). After voiding their urine, subjects dressed in only light clothing were weighed on a balance scale (model 2020 Toledo scale; Toledo, Windsor, Canada) to the nearest 0.1 kg in the morning before each dietary period and on all experimental study days. Body mass index (BMI) was calculated as weight in kg divided by height in m2.
Isotope infusion studies
The stable isotopes used in these studies were NaH13CO3 (Cambridge Isotope Laboratories, Woburn, MA) with a 90% enrichment and L-[1-13C]phenylalanine (Tracer Technologies Inc, Somerville, MA) with a 99% enrichment at the carboxyl carbon. Quality-control tests performed by Tracer Technologies on the tracer [13C]phenylalanine showed a single spot by thin-layer chromatography, 99% enrichment at the specified carboxyl carbon position by nuclear magnetic resonance and gas chromatography with mass spectrometry, and absence of the D-isomer confirmed by gas chromatography with a chiral column. Stock solutions of NaH13CO3 (3.30 g/L), priming L-[1-13C]phenylalanine (6.65 g/L), and constant-infusion L-[1-13C]phenylalanine (3.00 g/L) were prepared in normal saline solution. The solutions were prepared with sterile water by passage through a 0.22-µm filter (Millipore, Bedford, MA) under a laminar flow hood and were then dispensed into single-dose vials in the Department of Pharmacy of The Hospital for Sick Children. Each batch of the infusates was shown to be sterile and free of bacterial growth over 7 d in culture and to be pyrogen-free by the limulus amebocyte lysate test (21).
Isotope infusion studies were carried out on days 3, 6, and 9 of the study period in a temperature-controlled metabolic facility at The Hospital for Sick Children. The meals at 0800 and 1200 were divided into 6 equal parts, which the subjects consumed hourly beginning 2 h before infusion of the isotope (18); this ensured a metabolic steady state. The amount of dietary phenylalanine in the last 4 meals was reduced by a quantity equivalent to the amount of L-[1-13C]phenylalanine tracer infused so that the total phenylalanine intake remained unchanged. We used L-alanine to keep the total amino acid content constant, decreasing its amount as threonine intake increased. The test threonine intake and the nonlabeled phenylalanine were provided separately on study days so that their exact intakes would be known. Three baseline breath and blood samples were obtained before infusion of the isotope, which was infused via a 21-gauge peripheral catheter inserted aseptically into the antecubital fossa vein of the left arm. Each subject received a priming dose of NaH13CO3 (0.176 mg/kg) and a priming dose of L-[1-13C]phenylalanine (0.665 mg/kg). Immediately thereafter, a constant infusion of L-[1-13C]phenylalanine (1.2 mgkg-1h-1) was administered for 4 h via a calibrated syringe pump (IVAC 710; Eli Lilly Inc, San Diego).
Once the isotope infusion started, blood and breath samples were taken every 30 min during the 120240 min after priming, when steady state was achieved. Arterialized venous blood (3 mL) was sampled from a 21-gauge peripheral catheter inserted into a superficial vein on the dorsum of the right hand; the catheter was kept patent by infusing the line with heparin (10000 US Pharmacopoeia U/L) between sampling. To arterialize the venous blood, the hand was heated inside a heating device at 60°C for
15 min before blood sampling (22). Blood samples were drawn into heparin-containing syringes and stored on ice until centrifugation at 1500 x g for 10 min at 4°C. Plasma was stored at -20°C until it was analyzed for [13C]phenylalanine enrichment.
Three 7-min baseline breath samples were collected 1 h before the infusions started. Thereafter, 7-min breath samples were collected every 30 min during the 4-h infusion period. The collection of breath samples and the procedure for trapping expired carbon dioxide were reported in detail previously (23). In addition to the usual breath-sample collection, indirect calorimetry (Sensormedics 2900; Sensormedics) was carried out to measure the total carbon dioxide production rate simultaneously with breath sampling both before and every 30 min throughout the isotope infusion. A ventilated face mask (Scott 80216730; Sensormedics) was used.
Analytic procedures
The percentage enrichment of the expired 13CO2 was measured on a dual-inlet magnetic sector isotope ratio mass spectrometer (VG Micromass 602D; VG Isogas Ltd, Cheshire, England) by using techniques described previously (24). Breath samples were analyzed in duplicate. Breath enrichments from baseline samples and from those taken during the isotope infusion were expressed as atoms percent excess (APE) 13CO2 over a reference standard of compressed carbon dioxide gas.
Plasma samples (200 µL) were deproteinized and acidified with an equal volume of 20% trichloroacetic acid (wt:vol) and were then centrifuged at 9000 x g for 5 min at room temperature. Amino acids were separated from the supernate by a cation exchange resin (Dowex 50W-X8, 100200 mesh H+ form; Bio-Rad Laboratories, Richmond, CA) and were derivatized to their N-heptafluorobutyryl n-propyl esters by using the method described by Patterson et al (25). Amino acid enrichment was measured on a gas chromatograph (Hewlett-Packard model 5890, series II; Hewlett-Packard, Mississauga, Canada) attached to a quadrupole mass spectrometer (VG Trio-2; VG Isogas Ltd). Separation of the amino acid derivatives was performed with helium as the carrier gas on a 30-m x 0.32-mm (inside diameter) x 1.0-µm (film thickness) fused silica capillary column (HP5; Hewlett-Packard) coupled directly to the ion source, which was operated under conditions of negative chemical ionization with ammonia as the reactant gas. Each amino acid was analyzed by splitless injection on an automatic sampler (HP 7673 injector; Hewlett-Packard). Selected ion chromatographs were obtained by monitoring the mass-to-charge ratios of 383 and 384 for [13C]phenylalanine, corresponding to the unenriched (m) and enriched (m+1) peaks, respectively. The areas under the peaks were integrated by a Digital DECp 450D2LP computer (Digital Instruments, Santa Barbara, CA) with the LAB-BASE program (VG Isogas Ltd).
Data analysis
The model used here to study phenylalanine metabolism was described previously by researchers who used a constant infusion approach to study amino acid oxidation (26). The overall model was as follows:
![]() |
![]() |
|
Statistical analysis
A two-factor (threonine intake and subject) general linear model (28) was used to assess the relations among the rate of release of 13CO2, phenylalanine flux, and phenylalanine oxidation. Nonsignificant interactions (P
0.1), such as the order in which the test intakes were given, were dropped from the model to produce the final reduced model. The significance of specific differences between groups was examined by using the least-significant-differences multiple range test.
The mean requirement for threonine was determined by breakpoint analysis with a two-phase linear regression crossover model (29) similar to that described previously (17, 20). This model minimizes the residual SE in a stepwise partitioning of data points between 2 regression lines. The 95% CI for the mean threonine requirement was calculated by Fieller's theorem (28) and the upper number of the 95% CI was considered the safe amount for threonine intake. In all cases, results were considered to be significant at P
0.05.
| RESULTS |
|---|
|
|
|---|
Two-factor general linear model results revealed that both subject (P = 0.002) and threonine intake (P = 0.012) had significant effects on the rate of release of 13CO2 (F13CO2) from oxidation of tracer [1-13C]phenylalanine. Although the subjects differed significantly in their absolute F13CO2, they showed a similar pattern in which F13CO2 decreased with increasing threonine intake to a point that was close to the requirement, after which it remained relatively stable. The mean (±SEM) values for F13CO2 (n = 6) at threonine intakes ranging from 5 to 35 mgkg-1d-1 are shown in Figure 2
. Regression analysis showed that phenylalanine catabolism, as measured by F13CO2, decreased as threonine intake increased from 5 to 20 mgkg-1d-1. By using a two-phase linear regression crossover model, a breakpoint in the F13CO2 response curve was identified at a dietary threonine intake of 19.0 mgkg-1d-1; the upper number of the 95% CI was 26.2 mgkg-1d-1. The threonine requirements of the individual subjects were estimated by visual inspection and were found to be 35, 20, 20, 10, 20, and 20 mgkg-1d-1 (Figure 3
).
|
|
| DISCUSSION |
|---|
|
|
|---|
Attempts to determine total threonine oxidation by using the direct amino acid oxidation method were first made in mammals; researchers measured the production of labeled carbon dioxide after infusion of threonine with either radioisotope or stable isotope labeling of the carboxyl group (3134). However, these studies failed to account for a major confounder, the sequestration of carbons 1 and 2 of threonine into glycine via the threonine dehydrogenase (TDG) pathway (16). Ballevre et al (16) developed a multitracer method to quantify both major pathways of threonine oxidation in pigs and found a partition in which 80% of oxidation occurred via the TDG pathway and 20% occurred via the threonine dehydratase (TDH) pathway. Furthermore, any increase in oxidation occurred almost exclusively via the TDG pathway (16). This effect of diet in pigs was confirmed by Le Floch et al (15), who also found evidence of extrahepatic oxidation of threonine associated with the decrease in total oxidation when the dietary supply of threonine was reduced. The same group later showed that in pigs, the pancreas is a major site of oxidation of threonine to glycine (35). The relative contributions of the 2 pathways to oxidation of threonine in humans have only been determined in one physiologic state. Darling et al (36) showed conversion of threonine to glycine, presumably via the TDG pathway, in low-birth-weight infants. They estimated that the rate of conversion of threonine to glycine in formula-fed preterm infants amounted to 44% of total threonine degradation, whereas degradation via the TDH pathway accounted for 56% of threonine disposal (36).
The only direct amino acid oxidation study that estimated the threonine requirement in humans was performed by Zhao et al (14). Threonine oxidation rates fell with reduced threonine intake in 8 healthy young men, reaching a relatively constant rate between 10 and 20 mgkg-1d-1. This figure is 23-fold higher than the current recommendation (1). However, Zhao et al (14) found that excess threonine was not as efficiently oxidized as lysine (37) or leucine (38). They concluded that this could not be the result of alternative oxidative pathways because they could not measure any [13C]glycine. However, research in human infants (36) and in pigs (16, 39) showed partition of threonine metabolism to glycine via TDG. The analytic method used by Zhao et al (14) to measure [13C]glycine (gas chromatographyquadrupole mass spectrometry) was not sensitive enough to detect labeling, as compared with the recently developed method known as gas chromatographycombustion isotope ratio mass spectrometry (36).
IAAO is a new, independent, and relatively simple method of estimating amino acid requirements in humans. It has been validated extensively in animals (17, 20, 40). This technique is based on the hypothesis that the partition of any indispensable amino acid between oxidation and protein synthesis is sensitive to the amount of the most limiting amino acid in the diet. When an indispensable amino acid is limiting for protein synthesis, all other amino acids are in relative excess and therefore must be oxidized. This means that increasing the dietary amount of the limiting amino acid in graded quantities will increase the uptake of all dietary amino acids for protein synthesis, which in turn reduces their oxidation, until the requirement point is reached. Once the requirement is reached, further increments in the test amino acid will have no effect on the uptake of other indispensable amino acids for protein synthesis or oxidation (17).
We showed previously that phenylalanine oxidation is not altered by adaptation, over 3, 6, or 9 d, to phenylalanine intakes of either 4.2 or 14.0 mgkg-1d-1. Because the duration of adaptation for most amino acid requirement studies was 710 d, we concluded that the need for prior adaptation is minimal (4, 8, 9). These minimal requirements for adaptation allow for repeated measurements with each subject, and therefore data can be obtained on each subject across the full study range (4, 911). Nevertheless, it still remains to be determined whether subjects would show differences if they had adapted to lower intakes of threonine for a period of several weeks.
The precursor pool from which phenylalanine is oxidized is a concern. This is the second study (see reference 11) in which we have not been able to obtain a breakpoint estimate from phenylalanine oxidation. Our estimate of phenylalanine oxidation was calculated on the basis of plasma phenylalanine enrichment, which may not represent enrichment at the site of oxidation within the hepatocyte. The actual precursor is [13C]tyrosine, and because its enrichment is likely to be lower than our measured value for plasma phenylalanine, our estimate of phenylalanine oxidation is lower than the true value. However, because we were able to achieve an isotopic steady state in breath 13CO2 enrichments, we reasoned that the rate of 13CO2 production varies in proportion to the rate of conversion of [13C]phenylalanine to [13C]tyrosine, and subsequent [13C]tyrosine oxidation. This in turn means that changes observed in the rate of 13CO2 production are a valid reflection of the use of phenylalanine for protein synthesis and hence permits a valid estimate of the requirement for the test amino acid (in this case, threonine). There was more individual variation in F13CO2 in this study (Figure 3
) than we observed in an earlier indicator oxidation study that was performed to determine tryptophan requirements (10). The reasons are not clear, except that the previously mentioned complex degradative pathways for threonine (15, 16) may result in more individual variation. The effect of graded amounts of threonine on the rates of threonine disposal via the TDG and TDH pathways has yet to be studied in humans.
Nevertheless, requirements for indispensable amino acids with more complex metabolism can be measured by IAAO. There is no dietary restriction on test amino acid intake that can be examined, because the requirement for the test amino acid is determined by the oxidation of a different amino acid. All possible dietary quantities of an amino acid can be studied, unlike the direct oxidation method, in which very low intakes are impossible because the amount of isotope infused has to be included as part of the dietary intake. The statistical estimates of breakpoint are also stronger when there is a wider range of intakes of the test amino acid, both below and above the inflection point. We published previously a review of studies that used IAAO to evaluate amino acid requirements in humans (41). In the current experiment, we found a mean threonine requirement of 19.0 mgkg-1d-1, with an upper safe amount of 26.2 mgkg-1d-1. This latter amount is almost 4-fold higher than the current recommended threonine requirement of 7 mgkg-1d-1 (1). The mean IAAO-determined values are marginally (2225%) higher than the direct oxidation estimate of 1020 mgkg-1d-1 (14) and Millward's (6) recalculated nitrogen balance estimate of 15.7 mgkg-1d-1, whereas the safe amount is 6070% higher. These observations provide further evidence that the current dietary recommendations for indispensable amino acids require revision.
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. L. Urschel, M. Rafii, P. B. Pencharz, and R. O. Ball A multitracer stable isotope quantification of the effects of arginine intake on whole body arginine metabolism in neonatal piglets Am J Physiol Endocrinol Metab, September 1, 2007; 293(3): E811 - E818. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Elango, M. A Humayun, R. O Ball, and P. B Pencharz Lysine requirement of healthy school-age children determined by the indicator amino acid oxidation method Am. J. Clinical Nutrition, August 1, 2007; 86(2): 360 - 365. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. E. Matthews An Overview of Phenylalanine and Tyrosine Kinetics in Humans J. Nutr., June 1, 2007; 137(6): 1549S - 1555S. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W-C Hsu, L. A Goonewardene, M. Rafii, R. O Ball, and P. B Pencharz Aromatic amino acid requirements in healthy men measured by indicator amino acid oxidation Am. J. Clinical Nutrition, January 1, 2006; 83(1): 82 - 88. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. J Moughan, C. A Butts, A. M Rowan, and A. Deglaire Dietary peptides increase endogenous amino acid losses from the gut in adults Am. J. Clinical Nutrition, June 1, 2005; 81(6): 1359 - 1365. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. J. Reeds and P. J. Garlick Protein and Amino Acid Requirements and the Composition of Complementary Foods J. Nutr., September 1, 2003; 133(9): 2953S - 2961. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. V Kurpad, T. Raj, M. M Regan, J. Vasudevan, B. Caszo, D. Nazareth, J. Gnanou, and V. R Young Threonine requirements of healthy Indian men, measured by a 24-h indicator amino acid oxidation and balance technique Am. J. Clinical Nutrition, October 1, 2002; 76(4): 789 - 797. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. V Kurpad, M. M Regan, T. Raj, A. El-Khoury, R. Kuriyan, M. Vaz, D. Chandakudlu, V. G Venkataswamy, S. Borgonha, and V. R Young Lysine requirements of healthy adult Indian subjects receiving long-term feeding, measured with a 24-h indicator amino acid oxidation and balance technique Am. J. Clinical Nutrition, August 1, 2002; 76(2): 404 - 412. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Kriengsinyos, L. J. Wykes, R. O. Ball, and P. B. Pencharz Oral and Intravenous Tracer Protocols of the Indicator Amino Acid Oxidation Method Provide the Same Estimate of the Lysine Requirement in Healthy Men J. Nutr., August 1, 2002; 132(8): 2251 - 2257. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Borgonha, M. M Regan, S.-H. Oh, M. Condon, and V. R Young Threonine requirement of healthy adults, derived with a 24-h indicator amino acid balance technique Am. J. Clinical Nutrition, April 1, 2002; 75(4): 698 - 704. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Di Buono, L. J Wykes, R. O Ball, and P. B Pencharz Total sulfur amino acid requirement in young men as determined by indicator amino acid oxidation with L-[1- Am. J. Clinical Nutrition, December 1, 2001; 74(6): 756 - 760. [Abstract] [Full Text] |
||||
![]() |
P. B. Darling, J. Grunow, M. Rafii, S. Brookes, R. O. Ball, and P. B. Pencharz Threonine dehydrogenase is a minor degradative pathway of threonine catabolism in adult humans Am J Physiol Endocrinol Metab, May 1, 2000; 278(5): E877 - E884. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |