Mineral supplementation of white wheat flour is necessary to maintain adequate mineral status and bone characteristics in rats

.


Introduction
More than 50% of the world's total food energy is supplied by the eight most common grains species (wheat, rice, corn, oat, rye, barley, millet and sorghum).They are the major source of starch together with fibers, and they also contain non-negligible amounts of proteins, minerals and vitamins (1).The grains consumed in the industrialized countries are usually subjected to various types of process ing: milling, heat extraction, cooking, or other techniques which may alter the nutritional value of foods.Milling is a process that consists in separating bran and germ from the starchy endosperm in order to produce white flour.Howev er, such a refining results in significant nutrient losses, in particular fibers, minerals and vitamins.
Cereal products contribute substantially to the dietary supply of minerals, but since 60% of the minerals are pre sent in the aleurone cells, refined white flours are drasti cally depleted in minerals, compared to whole flours.Moreover, mineral bioavailability could be affected by the presence of large amounts of phytate, 85% of which is localized in the aleurone layer (2).This molecule is a strong chelator of mineral ions such as Ca 2 +, Zn 2 + and Fe 3 + and is not readily absorbed or digested in the small intes tine due to the low activity of intestinal phytase in humans (3)(4).The advantage of consuming whole cereals rather than refined cereals has often been questioned (5)(6).The potentially anti-nutritive effects of phytic acid have been mostly studied using diets supplemented to the same level of minerals (7)(8).
Except for calcium and sodium, a cereal like wheat can fulfill any mineral requirement but it is rich in fibers and phytic acid (0.32% of the whole grain) (1).It was thus of interest to compare the actual mineral bioavailability in rats fed diets in which the cereal product supplied was either white wheat flour supplemented or not in minerals, or whole wheat flour without any mineral supplement except for calcium carbonate and sodium chloride.The present study indicates that whole wheat flour represents a good source of available minerals, equal to that from the refined white flour supplemented in minerals.

Materials and reagents
Wheat flours were purchased from a local Massagette Mill, Massagette, France.Casein was purchased from Louis Franc;ois, St Maur, France.CaC0 3 was purchased from Sigma Chemical, St Louis, MO.NaCl was purchased from Merck, Darmstadt, Germany.Peanut oil was purchased from Bren dani, Clermont-Ferrand, France.All other chemicals were of the highest quality available.Distilled water was used throughout.

Animals and diets
Forty male Wistar rats (IFFA-CREDO, L'.Arbresle, France) that initially weighed 160 g were adapted to four experimental purified diets distributed as a moistened powder for six weeks.The control diet contained (g/kg) wheat starch : 650, casein : 200, peanut oil : 50, fibers (cellulose): 50, mineral mixture: 35, vitamin mixture: 10, DL-methionine : 3, Choline bitartrate : 2. The one whole flour and the two white flour diets contained 87 .5% of a French variety (Soisson, Soft Red Winter class) wheat flour (white or whole), 6% casein, 5% peanut oil, and 0.75% CaC0 3 , and 0. 7% NaCl.One of the two white flour diets was supple mented with Mg, Fe, Zn and Cu to reach the levels of min erals present in the whole flour diet.The mineral composi tion of these diets is shown in the Table 1.No vitamins were added because wheat flour provided sufficient amounts to meet essential requirements (9).The rats were housed two per cage (wire bottomed to limit coprophagy) and maintained in controlled temperature rooms (22°) with the dark period from 20.00 to 08.00.Food consumption and growth rate were measured twice weekly.During the last five days, the rats were transferred to metabolic cages (one rat per cage) and the feces and urine of each rat were collected.The rats were maintained and handled in accor dance with the recommendations of the Institutional Ethics Committee of the INRA (Clermont-Ferrand, France).

Sampling procedures
Rats were killed at the end of the dark period (09.00).They were anaesthetized with sodium pentobarbital ( 40 mg/kg) and maintained on a hot plate at 37°.Blood was drawn into heparinized syringes from the abdominal aorta (5-10 ml).Liver, femur and tibia were also sampled.Blood from each rat was placed in a plastic tube contain ing heparin and centrifuged at 10 OOOg for 2 min.After centrifugation, plasma was removed and kept at 4 ° for analysis.After blood sampling, the cecum with content was removed and weighed (total cecal weight).Cecal con tents were transferred into 2 microfuge tubes; one was

Analytical procedures
Phytic acid level was measured in the flours by a colori metric procedure based on the reaction between ferric ion and sulfos .alicylic aci�, as described by Latta & Eskin (10).Short Cham Fatty Acids (SCFA) were measured by gas-liq uid chromatography on an aliquot of supernatants of cecal contents (20,000 x g, 10 min at 4°) (11).The SCFA cecal pool was calculated as follows: Cecal pool (mmol) =cecal concentration (µmoljl) x cecal water content weight (g).Transferrin saturation percent and plasma iron were deter mined using Ferrimat-Kit and TIBC additif from Bio Merieux (Lyon, France).
Mg and Ca were determined after mineralization of tis sues, food and feces samples (10 h at 500°).The resulting residues were extracted with 5 moljl HCl and made up to an appropriate volume with 1 g/l lanthanum chloride solution.To determine Fe, Zn and Cu levels in liver, tibia, food and feces, 0.25 to 0.5g of dried samples were dry-ached (10 h at 500°) and then extracted at 130° in HNO /H O (2/1) (Merck, Suprapur) until discoloration; final �ilu\i6n was made in 0.5 M HN0 3 • Ca and Mg concentrations were mea sured after a 50-fold dilution of plasma in lanthanum chlo ride solution.the other trace elements were measured after a 5-fold plas m a dilution in HCl and tri-chloroacetic acid for protein elimination.Mineral concentrations were deter mined by atomic absorption spectrophotometry (Perkin Elmer 420, Norwalk, CT) in an acetylene-air flame at the following wavelengths : 422 (Ca), 285 (Mg), 248 (Fe), 214 (Zn) and 325 (Cu).A nebuliser with high sensitivity was used for trace element determinations.Appropriate quality controls were run with each set of measurements.

Determination of bone characteristics
Femoral mechanical testing: The mechanical resistance of femoral bones was determined using a three point bend ing test as previously described (12).Thus, the femoral failure load and Young modulus were determined.
Bone mineral density: Dual energy X-ray absorbiometry (DEXA) measurements were made with a Hologic QDR-4500 A X-ray bone densitometer (Hologic France, Massy, France).Total femoral bone mineral density (TBMD) was deter mined.In addition, the bone mineral densities of two sub regions, one corresponding to the proximal metaphyseal zone (MBMD) which is rich in calcaneus bone, and the other to diaphyseal zone (DBMD) which is rich in cortical bone, were also measured as previously described (13).

Statistical analysis
Standard procedures were used to calculate means and stan dard deviation (SD).Results were compared by one-way analysis of variance (ANOVA) using Instate software.The analysis of variance test was followed by a student-New man-Keuls multiple comparisons test.Because ANOVA test assumes that standard deviations are equal, the non para metric Kruskal-Wallis test was applied when the difference among standard deviations is significant.Differences were considered as significant between groups when p < 0.05.

Diet analyses and animal growth
The chemical analysis showed that the whole flour diet � ompared to the white flour diet was almost 3-fold highe� rn Mg, and about 100% higher in Zn, Fe, and Cu.Conse � uently, the daily supply in Mg, Fe, Cu and Zn was higher rn rats fed whole flour than in those fed white flour, since both food intake and mineral density were higher in rats fed the former diet.The supplementation of white flour wi .
th Mg, Fe, Zn and Cu improved the daily supply of these minerals, to meet the dietary recommendation intakes of minerals in rats.
As shown in figure 1, body growth was faster in animals receiving the control diet than in the other animals.Rats receiving the white flour diet had a significantly slower growth rate compared to the other groups.Moreover, the white flour diet was consumed in significantly smaller quantity and the weight gain of rats fed this diet was sig nificantly lower (-38%) than that of rats fed the whole flour diet (table 2).However, no clinical deficiency sign was noted in rats fed the white flour diet.

Fecal and cecal variables
As expected, white flour ingestion, supplemented or not in minerals, was accompanied by a significant decrease in the amount of feces compared to whole flour or control diets groups (results not shown).In addition, the cecal SCFA pool was significantly higher in the whole wheat flour group compared to white flour group, resulting in a significantly lower cecal pH.However, the cecal content and cecal wall weights were not significantly different between whole flour and white flour diet groups.The min eral supplementation of white flour markedly increased the cecal content weight and the cecal SCFA pool, com pared to the white flour group (Table 2).

Bone characteristics
White flour ingestion elicited a significantly lesser bone resistance (about -25%), proximal metaphyseal and dia physeal bone mineral densities and total bone mineral density, in comparison to rats fed the control or the whole  3).The supplementation of white flour with minerals was efficacious to normalize the bone char acteristics, which became statistically similar to those observed in the control and the whole flour groups.

Ca and Mg status
The Ca level was similar in the three experimental diets, but 20% higher in the control diet.Consequently, Ca sta tus (urinary and bone Ca) did not differ in the different groups, and no negative effect of the whole flour was noted.However, significantly greater plasma Ca level was observed in the white flour group compared to the other diet groups.Paradoxically, tibia dry weight and its mechanical resistance were significantly diminished in the white flour group, in comparison with the other groups (tables 3,4).Mg intake was 80% lower in rats fed the white flour diet than in those fed the other diets.Consequently, all Mg markers (plasma, RB(, urine and tibia) were significantly lower in the rats fed the white flour diet for 6 wk compared to other groups.The mineral supplementation of white flour with Mg significantly improved the Mg status to that of the control or the whole flour groups (Table 4).

Trace element status
The daily Fe intake from white flour was half of that from whole flour.Consequently, plasma Fe and transferrin satu ration percentage as well as Fe tissue levels (liver and tibia) in rats fed white flour diet were twice as low as in those adapted to whole flour or control diets (table 5).
The mineral supplementation of white flour with Fe signif icantly improved the Fe status, to reach that of the con trol or the whole flour groups (Table 5).Finally, Zn or Cu intakes were more than double in rats consuming the whole flour diet, compared to the white flour diet.The levels of Zn in the plasma, in RBC and in liver were not significantly different among the four experimental groups (Table 5).However, urine and tibia levels of Zn were significantly higher in the white flour groups than in the control or the whole flour groups.Plas ma Cu level was slightly but significantly lower in the mineral supplemented white flour group compared to con trol or whole flour diets groups, whereas liver Cu level was not significantly different among the four experimental groups (Table 5).

Discussion
The literature concerning the impact of whole flour, rich in dietary fibers, on mineral bioavailability is discordant.Negative effects on mineral bioavailability have been fre quently ascribed to foods rich in dietary fibers or their associated compounds (3,5,8,14 ).This is why many nutritional recommendations advise against intake of large amounts of unrefined cereal products.Consequently, the grains consumed in the industrialized countries are usually subjected to various types of processing, in partic ular the milling process.Unfortunately, such a refining results in substantial nutrient losses (fibers, vitamins and   .For example, the mineral level in the whole wheat flour used in the present study was 1 to 2 fold higher than in the refined white wheat flour.Thus, it would be important to evaluate during a long period (sev eral months) the impact of such refined products on min eral status and to check whether unrefined products can modify this status, especially when these products repre sent an important source of minerals.
The diets used in the present study differed in several respects including fiber, phytic acid and mineral concen trations (except for Ca and Na).The experimental diets ensured a well balanced supply of proteins, lipids and car bohydrates.The whole flour and the mineral supplement ed white flour diets met several of the American Institute of Nutrition (AIN) mineral recommendations whereas the mineral levels in the white wheat flour diet were largely lower than the recommended levels.
The results of the present experiment showed clearly that the ingestion of white flour, deficient in minerals, led to overt deficiency in Mg and Fe.We also showed that the supplementation of the white flour with minerals was effi cacious to improve the mineral status which therefore attained that observed in the control or the whole flour diet groups.Altogether, this experiment showed that unrefined flour may be a good source of minerals, equal to that from refined flour supplemented with minerals.More over, it must be kept in mind that numerous other func tionally active constituents are present in the unrefined flour.
Because wheat flours are generally poor in Na and Ca, the three experimental flour diets were supplemented with these cations.Ca intake was thus similar in the four experimental groups.In spite of the differences in the characteristics of the tested flours, the urine and bone  levels of Ca were similar in the four studied groups.How ever, hypercalcaemia was observed in rats receiving the white flour diet.This hypercalcaemia may be simply the result of the severe Mg deficiency which is known to induce an hypercalcaemia in rats (15).The fact that Ca status remained unchanged in rats receiving the whole flour diet or the mineral supplemented white flour diet is not surprising, since the phytic acid/Ca ratio was low (0.23).The bone Ca level confirms that the use of whole flour diet had no negative effects on Ca metabolism compared to the control or to the mineral supplemented white flour diet groups.In a previous study, it has been shown that Ca absorption from whole or white flour diets containing the same levels of Ca was not different in rats (16).
The results of the present study show clearly that white flour ingestion accompanies with a severe Mg deficiency, whereas whole flour, rich in dietary fibers and in Mg, appeared to be a very interesting source of available Mg, ensuring a good Mg status reflected by both plasma and tibia Mg levels.These results are in good agreement with a previous study showing that Mg absorption was better from whole flour than white flour diets in rats (14).The supplementation of white flour with Mg normalized the Mg status (plasma, RBC and urine) which became similar to that observed in the control and the whole flour groups.Whole flour was not less efficient on Mg status than supplemented white flour.The principal route of Mg absorption in the distal part of the digestive tract is the passive para-cellular pathway whereas active trans-cellu lar transport is relatively unimportant (17).Thus, concen tration and solubility of Mg and intestinal epithelium per meability are probably the principal determinants of Mg absorption (18).Cereal products are good sources of Mg, and a slightly acidic pH in the cecum can promote an effective solubilization of Mg (19).
Calcium and Mg, together with phosphorus are the major constituents of bone ash.The bone quality is directly related to the amount and the bioavailability of these minerals in the diet.In the present study, the deter mination of femoral characteristics on the left femur of rat indicated that the white flour ingestion decreased sig nificantly the bone mineral density which is an important factor in determining bone strength and known to be strongly associated with fracture risk.The unaltered Ca status may appear in disagreement with the femoral bone characteristics in the studied groups.Indeed, these bone characteristics are dependent on many factors such Ca status, dietary P-Ca ratio and status of the other dietary minerals, in particular Mg.Mg participates in the normal formation and remodeling of bone and its status influence the biomechanical function of bone (20).60% of Mg of the organism is concentrated in the bone and hence a severe deficiency in Mg may seriously negatively affect the bone characteristics.The supplementation of white flour with minerals, especially Mg, allowed to normalize the bone characteristics which became close to those observed in animals fed control and whole flour diets.The effects of the supplemented white flour and the whole flour were not significantly different, which means that whole flour ingestion was without negative effects on bone characteristics in rats, in contrast to white flour ingestion.
The daily intake of Fe, Zn and Cu was significantly lower in rats fed the white flour diet than in those fed the whole flour diet.Supplementation of white flour diet with minerals afforded a mineral intake close to that observed in control or whole flour diet groups.The whole flour diet contained twice more Fe than the white flour diet did, translated in a significant decrease in Fe status with the white flour diet ingestion.Indeed, the present results indicate that rats fed the white flour diet have plasma, liver and tibia concentrations of Fe twice lower than those found in rats fed whole flour.Rats fed the white flour diets showed also a significant decrease in hemoglobin level (-16%) in comparison to the rats fed whole flour, indicating an anemia in these rats.A good intestinal absorption of Fe in the lower part of the digestive tract could probably explain this effect, as a result of luminal pH lowering and micro flora activity (21).Indeed, in a previous study, Fe absorption was found more effective from whole flour than from white flour diets in rats (14).Because Fe homeostasis is controlled by absorption, an increase in Fe absorption by more than 100% may be responsible for this rise of plasma Fe in rats fed the whole flour diet.This is one unique aspect of Fe metabolism which takes place in a closed circuit.In fact, the Fe pool is subjected to a permanent renewal and this process is nearly independent of external Fe.Thus, Fe deficiency can develop when intake is dramatically low, and Fe overload can also occur when the daily absorbed amount of Fe is increased (22,23).The supplementation of white flour with Fe allowed to normalize the Fe status which became close to that observed in animals fed whole flour or con trol diets.
In spite the fact that rats fed for six weeks the white flour diet, 50% lower in Zn and Cu content than the other diets, the present results indicate that Zn or Cu status were not significantly modified in rats receiving the white flour diet in comparison to those of rats receiving the control or the whole flour diet.In addition, the supple mentation of white flour with Zn and Cu did not elicit any effect on these parameters.Generally, Zn and Cu home ostasis is well controlled, and remains constant over a wide range of dietary intake levels (24).Indeed, the endogenous excretion of Zn or Cu in the feces is very reg ulated according to the dietary intake.Only very low or very high dietary intakes for long periods cause the regu latory mechanisms to become overloaded resulting either in a depletion or an accumulation of the element in the body.Therefore, it could be interesting to study for a longer time the effect of such flours on Zn and Cu status.
Negative effects of dietary fibers and phytic acid by mineral complexing have been often studied by adding fibers or phytic acid directly in diets.Such an experimen tal protocol fails to take into account the mineral content of fiber-rich products, as in the present experimental diets.Such products are known to be highly rich in miner als (25).It is well established that rats have a high capacity to hydrolyze phytic acid, compared to humans who have very low intestinal phytase activity (26).How ever, when wheat is eaten in completely leavened bread, more than 50% of phytic acid is already destroyed (27).Furthermore, it is noteworthy that the maintain of the phytase activity in cereal products may allow partial hydrolysis of dietary phytate in the stomach, improving the bioavailability of minerals.
In conclusion, even if rat is not a perfect model to evaluate mineral bioavailability for man, the results of this experiment show clearly that whole flour was practi cally as effective on mineral status as white flour ade quately supplemented with minerals.The ingestion of unrefined products was not accompanied by negative effects on mineral status, and then can be encouraged.Moreover, recent epidemiological and clinical studies have shown interest in increasing the consumption of phytic acid-rich products in preventive nutrition.The possible negative effects of such products on mineral sta tus can be less important when these products are taken together with the other components of the meal.It is thus wise to promote the consumption of whole grains rather than of purified cereal products, to minimize the loss of functionally active constituents of grains during industrial processing and to optimize the mineral status in humans.Human studies are still needed to confirm these results in rats.

Figure 1 .
Figure 1.Effect of experimental diets on animal growth rate Values are given as mean ±SD, (n = 10).Experimental diets : A : Control; B : Whole wheat flour; C : White wheat flour supple mented with minerals; and D : White wheat flour

Table 2 .1
Daily weight gain, food intake, fecal excretion and fecal variables in rats fed for six weeks control or white wheat flour or whole wheat flour adjusted or not for minerals1,Values are given as mean ±SD, (n = 10).2Means in a row not sharing a superscript are significantly different (p<0.05).

Table 1 .
Fiber and phytic acid contents and mineral composition of experimental diets.
1 These flour diets were supplemented with calcium carbonate and sodium chloride.

Table 3 .
Bone characteristics in rats fed for six weeks control or whole wheat flour or white wheat flour adjusted or not for miner als1,2.

Table 5 .
Iron, zinc and copper concentrations in rats fed for six weeks control or whole wheat flour or white wheat flour adjusted or not for minerals1'2• 1 Values are given as means± SD (n = 10).2 Means in a row not sharing a superscript are significantly different (p<0.05).