Contents - Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Cyril O. Enwonwu
The sickle gene results from a point mutation, GTG GAG in the codon for the sixth amino acid of ▀-globin. Inheritance of this mutant gene elicits one of the sickling syndromes, either the heterozygous carrier state (sickle-cell trait, haemoglobin [Hb], as genotype [HbAs]) or one of the more severe conditions identified by the generic term sickle-cell disease (SCD). Sickling disorders are increasingly affecting members of many races and nationalities as a result of intermarriage [1; 2] and constitute the major genetic disease among blacks. It is currently estimated that about 70 million people world-wide carry the sickle cell trait, at least 75% of whom live in Africa, mainly south of the Sahara . Homozygous sickle-cell anaemia (HbSS), considered the most common and severe form of SCD, is variously estimated to occur in approximately 1 of 400 to 625 live births among the black population of the United States [2; 4].
The homozygous condition is characterized by markedly accelerated turnover rate of haemoglobins, which may be as high as 6 to 20 times normal [5; 6], increased susceptibility to severe bacterial infections , a haemolytic anaemia, delay in somatic growth and sexual maturation [2; 7], numerous vaso-occlusive events, and prominent deterioration in renal function with age [2; 8] among other features. Increased rates of tissue (Hb) destruction and repair, as well as frequent infections, commonly associated with HbSS, are among the important factors that modify whole-body protein turnover . Indeed, recent studies in children  and adults  with HbSS suggest almost a twofold increase in whole-body protein turnover, and this implies an equally marked increase in energy expenditure [9; 12-14]. Associated with the hypercatabolism of HbSS are high faecal and urinary losses of protein that are largely independent of intake levels [9; 12; 15]. These are indicative of increased requirements for dietary energy and protein in HbSS.
This study, using three-day food-intake records, changes in plasma amino acid patterns, 24-hour urinary excretion of some amino acids, and other relevant biochemical indices, was designed to examine the requirements for protein and amino acids in HbSS qualitatively, particularly with reference to the commonly used recommended dietary allowances (RDA) . The importance of the study also derives from the often-forgotten fact that haemoglobin is structurally an unbalanced protein totally devoid of Ile but very rich in Val, Leu, Gly, and His, thus raising the strong possibility that amino acids present in most diets may not be present in the correct proportions to meet the accelerated demand for haemoglobin synthesis in HbSS [12; 19; 20]. In addition, the study examined zinc loss in urine in relation to intake in HbSS, since zinc is closely linked with protein metabolism and particularly with the haemoglobin molecule and its turnover [7; 21; 22].
Materials and methods
The detailed protocol for this study was reviewed by the Human Investigation Committee, Meharry Medical College, Nashville, Tennessee, USA. All participants in the study gave their informed consent.
Afro-American United States adults with sickle-cell disease (HbSS) who receive their routine medical care at the Comprehensive Sickle Cell Center, Meharry Medical College, were enrolled in the study. Confirmation of HbSS was by haemoglobin electrophoresis on cellulose acetate and by family history. The subjects were free of overt infections, had not received any blood transfusion in the preceding three months, were not in a crisis state, and were not taking any medications except for occasional use of the common over-the-counter analgesics. The control subjects (HbAA) were matched for age and sex, were drawn from the same racial and socio-economic backgrounds as the HbSS group, and did not carry the sickle-cell gene.
Diet was assessed by examining three-day food-intake records immediately before anthropometry and collection of blood samples for biochemical studies. Using plastic food models and appropriate descriptions of food portions in common household measures, a registered dietitian instructed the subjects on how to keep accurate intake records. The data were analysed using the Nutripractor 6000 computer software designed by Practorcare (Practorcare, San Diego, Calif., USA). The amino acid contents of the foods consumed were estimated using standard food tables .
After an overnight (8- 12 hours) fast, blood was collected by venipuncture into evacuated heparinized polypropylene tubes. Non-heparinized blood was also collected into acid-washed vials and used for serum preparation. Plasma was separated from the heparinized aliquot within 0.5 hour of collection, deproteinized with 5-sulfosalicylic acid (50 mg SSA per millilitre of plasma) containing an internal standard (S-2-amino ethyl-L-cysteine), and centrifuged in the cold for 10 minutes at 5,000 g, and the deproteinized sample was either used immediately for analysis of free amino acid concentrations or stored at 70 ║-C for subsequent analysis.
Twenty-four-hour urine samples were collected in plastic jugs containing sodium fluoride tablets as a preservative, with the jugs refrigerated between collections. Collection of urine was begun after excretion of the first morning's sample, and included the next overnight sample. An aliquot of the 24-hour urine (200Ál), mixed with 400Ál of 15% SSA containing the internal standard (aminoethyl-L-cysteine, 225 nmol per millilitre), was centrifuged, the supernatant mixed with 400Ál of lithium-D buffer, and passed through a 0.22-mium Millipore filter. The filtrate (50Ál), containing the internal standard (50 nmol per millilitre), was used to estimate the levels of free ninhydrin-positive substances.
Free amino acids in plasma and urine were measured using the automatic amino acid analyser (Beckman model 6300 High-Performance Amino Acid Analyzer, Beckman Instruments, Palo Alto, Calif., USA), which was equipped with the Beckman Data System 7000. The analytical conditions were as described in the Beckman model 6300 operating manual. All the buffers and reagents used were commercial products from the Beckman Instruments Company.
Plasma level of the aromatic acid tryptophan was measured by the standard fluorimetric method of Denckla and Dewey as modified by Bloxam and Warren .
Estimation of plasma albumin concentration was by the dye-binding method involving use of the anionic dye bromocresol green (3,3',5,5'-tetrabromo-m-cresolsulfon-ptthalein) . Human serum albumin fraction V (Sigma Chemical, St. Louis, Mo., USA), corrected for moisture content, served as the standard.
Serum and urinary levels of zinc were determined by atomic absorption spectrophotometry  using the Perkin-Elmer Spectrophotometer, model 303 (Perkin-Elmer Corp., Norwalk, Conn., USA). The samples were deproteinized with 10% trichloroacetic acid.
Blood haemoglobin level (grams per 100 ml) was measured by optical density at 540 nm, using Drabkin's (cyanmethaemoglobin) reagent (Fisher Diagnostics, Orangeburg, N.Y., USA). Other red blood cell indices were measured according to standard procedures, using a Coulter Counter, model ZB1.
Orotic acid level in urine was determined by the colorimetric reaction of Rogers and Porter as modified by Harris and Oberholzer . Most of the interfering substances were removed on a small column of Amberlite CG/20-200 mesh cation-exchange resin (Sigma Chemical), and a control reaction was incorporated to correct for residual background colour . To eliminate any false positive results, pregnant individuals and those taking any purine analogues were not included in this section of the study . Colorimetric determination of creatinine at 500 nm was carried out using standard Sigma diagnostics (procedure no. 555, Sigma Chemical).
Differences between group means were tested by the Kruskal-Wallis non-parametric analysis of variance available on the SPSS (Statistical Package for the Social Sciences).
Table 1 summarizes several features of the HbAA controls and HbSS subjects at the time of this study. Energy, protein, and zinc intakes were comparable between the two groups and were within the RDA for healthy individuals [16; 18; 23]. Mean body weight, blood haemoglobin concentration, and haematocrit were significantly lower in the subjects with HbSS, while 24-hour urine volume was prominently greater (P < .05), than in the controls.
TABLE 1. Characteristics of the HbAA (control) and HbSS subjects at the time of the study
HbAA (N = 13)
HbSS (N = 22)
30 ▒ 5
29 ▒ 3
82.6 ▒ 6.4
62.1 ▒ 2.3
176.9 ▒ 4.9
169.5 ▒ 8.4
|Haemoglobin (g/100 ml)||
13.7 ▒ 1.5
7.5 ▒ 1.1
40.6 ▒ 1.2
20.9 ▒ 0.9
|Energy intake (kJ * kg-1 day-1)||
164.9 ▒ 8.1
156.7 ▒ 5.0
|Protein intake (g * kg-1 day-1)||
0.96 ▒ 0 37
1.17 ▒ 0.24
|Zinc intake (mg/day)||
10.99 ▒ 0.89
|Urine volume (L/24 hr) (a)||
1.06 ▒ 0.24
2.15 ▒ 0.55
Data expressed as means ▒ SEM
a. Derived from 7 subjects with HbAA and 12 with HbSS.
We could not detect any differences in intake levels of the indispensable and semi-essential amino acids between the groups (table 2). All subjects ingested the various amino acids at levels two to four times as high as the recommendations by the FAO/WHO/ UNU Expert Consultation  and also generally higher than the revised estimates of the requirements in adults derived from a more reliable kinetic approach [17; 18]. The ratio of the sum of indispensable amino acid intake to total protein intake was 32.96% in the HbAA controls and 28.32% in subjects with HbSS (see tables 1 and 2). The estimated intake of His by our study groups was twice the level recommended for adults by the FAO/WHO/UNU panel  and five times the level of about 4 mg per kilogram per day suggested as adequate for healthy adults by some investigators [18; 29]. As indicated in table 2, estimated intake of Arg (ma kg-1 day-1) was about 44 to 46 in the two groups, an observation consistent with calculated daily intake of this amino acid from usual United States diets .
The subjects with HbSS had a 21% lower total concentration of fasting plasma indispensable amino acids (including Tyr, His, and Arg) (EAA) than the HbAA group, with no difference in the dispensable amino acids (NAA), resulting in a change in the EAA:NAA ratio from 70.9% in the HbAA to 55.2% in the HbSS group (table 3). The amino acids most profoundly affected were Arg (-39%), His (-32%), Leu (-31%), Tyr ( - 28%), Trp ( - 26%), Val (-24%), and Phe ( - 24%). The plasma levels of Gly (+20%) and 112 Cys (+36%) were significantly higher, while that of Ala ( - 22%) was lower, in the HbSS than in the HbAA group. It is possible that the reduced levels of amino acids noted in our study might be slightly exaggerated because of some plasma volume expansion that is believed to occur in persons with HbSS .
TABLE 2. Estimated amino acid intake (ma kg-1 day-1) of HbAA and HbSS subjects compared with recommended levels
|HbAA (N= 4)||HbSS (N= 10)||FAO/WHO/UNUa||VRYb|
a. Ref. 16.
b. Ref. 18
c. Includes Cys.
d. Phe + Tyr.
e. Sum of amino acids.
f. Ratio of sumAA to protein intake shown in table 1.
In both groups, there was marked individual variation in 24-hour urinary excretion of amino acids, but, on the whole, the subjects with HbSS excreted less amino acids than those with HbAA (table 4) despite a twofold greater 24-hour urine volume in the former group (see table 1). The amino acid levels in 24-hour urine that were most reduced in the HbSS group compared with the controls were Arg ( - 72%), Gly ( - 53%), His (-46%), and Thr (-40%), while excretion of 1/2 Cys (+73%) was significantly higher.
It would thus appear that, with the possible exception of Phe, the lower plasma levels of amino acids observed in HbSS were not due to exaggerated urinary loss but rather to greater flux and use in the tissues. Gly, His, and Arg accounted for 27.7%, 14.2%, and 0.48% respectively of total 24-hour urinary excretion of amino acids in the controls. Comparable values for the subjects with HbSS were 17.5%, 10.3%, and 0.18%, an observation suggestive of more efficient renal conservation and/or greater metabolic use of these amino acids in HbSS.
No differences in plasma albumin concentration and urinary creatinine levels were noted between the groups (table 5). The same is true of serum zinc levels, which is at variance with some reported findings  but in agreement with others .
In contrast with serum zinc levels, 24-hour urinary zinc excretion was twice as high in the subjects with HbSS as in the controls. Similarly, the urinary level of orotic acid was 3.3 times as high in HbSS, and this was significant (p < .01).
TABLE 4. 24-hour urinary levels of free amino acids (miumol/ 24 hr)
HbAA (N = 7)
HbSS (N = 12)
|Valine||57.9 ▒ 9.3||45.1 ▒ 7.2 (78)|
|Isoleucine||38.8 ▒ 6 2||41.7 ▒ 4.3 (107)|
|Leucine||39.0 ▒ 6.9||31.9 ▒ 4.5(82)|
|Threonine||250.1 ▒ 52.5||149.4 ▒ 17.3(60)|
|Methionine||63.6 ▒ 9.0||49.6 ▒ 11.2(80)|
|Tyrosine||157.7 ▒ 21.2||160.0 ▒ 14.1(67)|
|Phenylalanine||87.0 ▒ 25.8||136.2 ▒ 55.2 (157)|
|Lysine||181.9▒ 34.3||222.4 ▒ 38.9 (122)|
|Histidine||938.1 ▒ 160.6||504.2 ▒ 37.1 (54)|
|Arginine||31.4 ▒ 4.2||8.8 ▒ 2.2 (28)|
|Glycine||1,831.2 ▒ 480.7||853.6 ▒ 108.4 (47)|
|Alanine||379.4 ▒ 104.2||298.7 ▒ 40.5 (79)|
|1/2 Cysteine||46.4 ▒ 7.0||80.4 ▒ 10.3 (173)|
Levels are expressed as means + SEM Values in
parentheses represent percentages of the HbAA control levels.
a. Total essential amino acids including Tyr His, and Arg.
b. Total non-essential amino acids (only those that are significantly different for HbAA and HbSS are included in the table).
* Significantly different (P<.05 or P<.01).
TABLE 5. Albumin, creatinine, zinc, and orotate levels
|HbAA (N = 7)||HbSS (N = 12)|
|(g/100 ml)||4.88 ▒ 0.09||4.76 ▒0.12|
|(miug/dl)||89.60 ▒ 6.12||82.94▒12.40|
|(g/24 hr)||1.65 ▒ 0.24||1.58▒0.37|
|(mg/24 hr)||0.99 ▒0.22||3.29▒0.46|
Data expressed as means +/- SEM.
* Significantly different (P<.05 or P<.01).
Dietary intakes of protein, energy, the indispensable amino acids, and zinc by the HbAA and HbSS subjects investigated in this study were similar and adequate for healthy individuals of the RDAs for healthy individuals are totally irrelevant to the metabolic needs of persons with HbSS.
With minor exceptions, such as the markedly lower plasma His concentration (see table 3), the differences in plasma free amino acids in the HbSS subjects were consistent with the syndrome of protein-energy malnutrition (deficiency) in humans, and different from findings in simple energy deficiency (undernutrition or marasmus), in which levels of both dispensable and indispensable amino acids are equally reduced [33; 34]. The significantly lower plasma level of Ala in HbSS was one indication of protein-energy deficiency in this group, since Ala is quantitatively a very important gluconeogenic amino acid whose concentration in plasma rises when protein is deficient but energy status is adequate . In such a situation, Ala is not used for gluconeogenesis and metabolized to urea. When both energy and protein are deficient, as in HbSS, however, the plasma Ala level drops [33; 3].
Additional evidence of an inadequate status of both protein and energy in the patients with HbSS was that, in spite of the distorted plasma aminogram, plasma albumin concentration in this group was not different from that in controls (see table 5). This was consistent with a recent observation , and could be attributed to the fact that hypoalbuminaemia does not occur in protein-energy malnutrition unless energy consumption is in excess of requirements while protein intake is inadequate .
Among the indispensable and semi-essential amino acids whose levels in plasma were most severely affected in HbSS were Val, Leu, His, and Arg. Val plus Leu constitutes about 20% and His about 8% of the haemoglobin molecule. In addition, inadequacy of His reduces the activity of erythrocyte aminolevulinic acid dehydratase (porphoblinogen synthase, EC 188.8.131.52), a key enzyme in haem biosynthesis, thus adversely affecting erythropoiesis [29; 36; 37]. It is evident, therefore, that, with a much higher demand for amino acids for synthesis of a specific protein such as haemoglobin with its structural peculiarities in terms of amino acid composition, persons with HbSS are especially susceptible to a dietary imbalance pattern of amino acids [9; 12].
We postulate that the depletion of Arg in HbSS can be attributed to the significantly greater production of urea both in absolute terms and in relation to dietary protein intake [9; 11; 12; 15]. Our finding with respect to Arg is in good accord with reported observations that, when amino acid degradation is enhanced after intake of a high-protein diet  or an increased rate of endogenous protein breakdown, as is often evident in HbSS, this dibasic amino acid becomes indispensable 
That endogenous synthesis of Arg plus its normal dietary intake is probably not sufficient to meet the metabolic demands for the amino acid in HbSS was reflected in the prominent increase in urinary orotic acid in our patients. Orotic aciduria is considered a sensitive index of Arg deficiency in animals  and occurs also in humans when there is a significant increase in whole-body protein flux , even though a recent study on a very limited number of subjects and for a short duration questions its applicability to humans .
HbSS is characterized by significantly increased urinary urea, ammonia, and urate protein as well as by increased faecal protein at all levels of protein intake [12; 15]. The prominent increase in urea production is accompanied by more marked hydrolysis of urea in the bowel than in healthy individuals, and thus the person with HbSS, consuming what appears to be a normal protein diet, usually conserves protein in the same way as healthy persons on a low-protein diet [13-14]. Our HbSS subjects demonstrated what appeared to be more efficient renal conservation of several amino acids than the controls. For example, Gly accounted for 9.6% and 7.4% of total plasma amino acids in the two groups respectively. The ratio of Gly to total 24-hour urinary amino acid excretion was 27.7% in the controls but only 17.5% in the HbSS individuals. The latter was consistent with an enhanced metabolic demand for Gly in HbSS. It was estimated that the demand for Gly for synthesis of haem in HbSS is on the order of 1 to 2 g per day , and the observation of increased urinary excretion of 5-oxoproline (pyroglutamic aciduria) suggests limited availability of Gly in this genetic disorder .
Several landmark studies [17; 18; 42] have highlighted several problems inherent in the accurate determination of the amino acid and protein requirements of healthy individuals. The situation is even more complicated in chronic disease states, and, as rightly underscored by Beaton , food constituents under these conditions may exert important influences that are not implicit within the biologic functions that underlie nutritional requirements. It is now clear that dietary needs for protein and amino acids are modulated by the status of whole-body protein turnover and the activity of pathways associated with the catabolism of amino acids .
In healthy human adults with a whole-body protein turnover rate of about 3 to 4 g per kilogram per day, minimum intakes of indispensable amino acids to balance loss through irreversible oxidation (assuming a 70% efficiency for amino acid retention) are predicted by Young et al. [17; 18] to be at least two to three times as high as the FAO/WHO/UNU recommendation , and certainly should be about 33% and not 11% of the safe protein intake level (0.75 g kg-1 day-1) recommended by the latter. Estimated protein intake in the HbSS group was 1.17 g day-1 kg-1, and the intake of indispensable plus semi-essential amino acids in this group compared favourably with predicted minimum intakes for healthy human adults . It is clear, however, from the data provided in the various tables that these levels were not enough to maintain adequate protein nutritional status in the HbSS subjects.
Recent studies of whole-body protein turnover in adults with HbSS  suggest that it is twice as high as in HbAA controls (HbAA [g day-1 kg-1]: protein flux 0.5 + 0.02, protein synthesis 3.2 + 0.2, protein degradation 2.8 + 0.2; HbSS: protein flux 0.9 + 0.08, protein synthesis 6.0 + 0.5, protein degradation 5.6 + 0.5). It would thus appear that the predicted minimum intakes of indispensable amino acids for healthy human adults [17; 18] may need to be almost doubled to meet the significantly increased metabolic demands of uncomplicated HbSS. It is extremely difficult if not impossible to identify specific levels of these components that will cover most individuals with HbSS in view of the marked clinical diversity and severity of this condition . For example, various degrees of renal involvement are encountered in some adults with HbSS , and each person may have his own requirements .
The present study also showed prominent hyperzincuria in subjects with HbSS, an observation consistent with reports from other laboratories [7; 21; 45]. Chronic haemolysis of erythrocytes, which are rich in zinc, accelerate renal zinc clearance, and there is evidence of impaired renal tubular handling of this micronutrient in HbSS . We are currently evaluating the potential relevance of increased plasma and urinary 1/2 cystine levels to the genesis of hyperzincuria in HbSS. Zinc deficiency resulting from chronic hyperzincuria may have some bearing on the anorexia, poor growth, delayed onset of puberty, and hypogonadism frequently associated with HbSS [7; 21].
Sickle-cell anaemia is a chronic condition, with higher metabolic demands for several macronutrients and micronutrients [19; 46]. Recognition of this has important ramifications in the provision of proper nutritional support for victims of this disorder. It is likely, for example, that the prominent metabolic insufficiency of Arg, which our laboratory is the first to report in HbSS, may have important relevance to the clinical severity of the disease. Barbul  reviewed the detailed biochemistry, physiology, and therapeutic implications of this amino acid.
L-Arg serves as the precursor for production of nitric oxide (NO), now considered by many [48; 49] to be synonymous with one of the endothelium-derived relaxing factors (EDRFs) that inhibit platelet aggregation as well as their adhesion to the endothelial cell surface while at the same time promoting the relaxation of vascular smooth muscle. This EDRF, derived from the terminal guanidino protein atoms of L-Arg, is therefore of some potential importance in maintaining blood vessel patency as well as a nonthrombogenic luminal surface [48-50]. It is therefore necessary to investigate the contribution of L-Arg deficiency to the increased adherence of red blood cells to the vascular endothelium [51; 52] as well as the enhanced platelet aggregation and activation [53; 54] usually encountered in severe clinical cases of HbSS.
The importance of the subject of this report is further underscored by the fact that the racial groups and nationalities predominantly afflicted by this disorder constitute the main bulk of the socioeconomically deprived societies in the world [2; 3]. Protein-energy deficiency in persons with HbSS is usually complicated by concurrent deficiencies of several important antioxidant micronutrients such as zinc, vitamin E, and ascorbic acid [19; 20]. Vitamin-E depletion, for example, leads to rapid emergence of the "senescent cell antigen" on the red cell membrane , and, when this occurs in subjects with HbSS, the macrophages recognize the red cells not only as structurally abnormal but also as prematurely "senescent" , thus intensifying the already accelerated turnover of red blood cells.
Contents - Previous - Next