![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
The Journal of Nutrition Vol. 128 No. 10 October 1998,
pp. 1681-1687
Nutrition Department, The Pennsylvania State University, University Park, PA 16802
 |
ABSTRACT |
|---|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
To assess whether the molar ratio of retinol-binding protein (RBP) to transthyretin (TTR) is of utility in detecting vitamin A (VA) deficiency during inflammation, we analyzed data from a rat model of endotoxin-induced inflammation and from a previously reported randomized, placebo-controlled trial of VA supplementation in children with acute measles. In rats, both marginal VA deficiency and inflammation were independent causes of low plasma RBP (two-way ANOVA, P < 0.001), whereas plasma TTR concentration was reduced only by inflammation (P < 0.001). The molar ratio of plasma RBP to TTR was reduced (by ~50%) only in rats with marginal VA deficiency and inflammation (two-way ANOVA interaction, P < 0.01). Serum retinol concentration, C-reactive protein (CRP, an indicator of inflammation) and the RBP:TTR molar ratio were determined in children with acute measles at baseline and 2 wk after subgroups received a placebo or a 210 µmol VA supplement. The ratio of RBP:TTR was selectively reduced in children in the placebo group with low plasma retinol (<0.35 µmol/L) and elevated CRP (>40 mg/L). In children with a low RBP:TTR molar ratio (<0.30) at baseline, the RBP:TTR ratio increased significantly 2 wk later only in the VA-treated subgroup. These analyses provide evidence that, because RBP is differentially reduced in comparison to TTR during VA deficiency, the combined determination of the concentrations of serum RBP and TTR may provide a promising means of detecting VA deficiency during inflammation.
KEY WORDS: acute phase response · C-reactive protein · hyporetinemia · rats · humans · vitamin A supplementation
The existence of vitamin A (VA)6 deficiency and marginal VA status, especially in preschool-age children, remains an important public health problem in many parts of the world (Underwood 1994 Although several methodologies have been developed to assess VA status, the performance of methods based on the determination of plasma retinol or retinol-binding protein (RBP) is affected negatively by the presence of infection or inflammation. In normal fasting plasma, >95% of total retinol present is bound to RBP (Blomhoff et al. 1991 We previously demonstrated that the hepatic synthesis of RBP, assessed by the levels of RBP mRNA and protein, is reduced within several hours of endotoxin administration in rats with adequate VA stores (Rosales et al. 1996b The experiments reported here were designed to test the hypothesis that the combined determination of plasma RBP and TTR may provide a useful method of assessing VA deficiency during inflammation. This premise was based on several observations. First, both VA deficiency and inflammation can lead to indistinguishable reductions in plasma retinol and/or RBP concentrations (Goodman 1984 Animals and experimental design.
All experimental protocols were in compliance with the Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee. The model of endotoxin-induced inflammation in VA-sufficient rats was described previously (Rosales et al. 1996b Plasma retinol, RBP and TTR determinations.
Plasma retinol was determined by high performance liquid chromatography (HPLC) using trimethylmethoxyphenyl-retinol as an internal standard (Ross 1986 Post-hoc analysis of serum RBP and TTR concentrations in children with measles.
To assess the relationship of VA status and inflammation to the RBP:TTR molar ratio in a human population, we conducted a secondary analysis of previously collected data (Rosales 1993 Statistical analysis.
Data are reported as means ± SEM, or otherwise indicated. Exploratory data analyses were performed using measures of central tendency and distribution of variables by treatment and time of evaluation. A two-tail P-value < 0.05 was considered significant, or otherwise indicated. Data from animal studies were analyzed using one-way ANOVA with modified least significant difference (MLSD) post-hoc test to assess significance when comparing multiple means and a two-way ANOVA to assess the main effects and the interaction between VA status and inflammation (Rosner 1986 Rat studies.
The response to LPS treatment was characterized by an increase in core body temperature within 6 h; however, none of the LPS-treated rats showed signs of lethargy or shivering. The concentrations of plasma retinol, RBP and TTR, and the RBP:TTR molar ratio in LPS- and saline-treated VA-sufficient, VA-marginal and VA-supplemented rats are summarized in Table 1. In rats without inflammation, plasma retinol concentration was reduced significantly in marginally VA-deficient rats compared with VA-sufficient and VA-supplemented rats. Plasma RBP paralleled this reduction in retinol, but its change was not as steep. In contrast to the differences of plasma retinol and RBP concentrations between marginal VA deficiency and after VA supplementation, plasma TTR concentration was not affected by VA intake.
Effect of measles infection on serum retinol and its transport proteins.
Of 260 Zambian children with acute measles infection reported previously (Rosales and Kjolhede 1994
Evaluation of the RBP:TTR ratio among children with measles.
During convalescence, measles patients who had received a placebo had, by definition, a higher risk of developing VA deficiency than did the VA-supplemented group. Therefore, if the hypothesized relationship of a low molar ratio of RBP:TTR during VA deficiency and inflammation is true, this ratio should be reduced in children with measles who received the placebo relative to the group that received VA. Children with the lowest serum retinol concentration and highest CRP concentration had significantly lower RBP:TTR ratios than other placebo-treated children (Table 4). In contrast, the RBP:TTR ratio did not differ significantly by CRP and retinol levels among VA-treated children.
The goal of this study was to assess the biological significance of the RBP:TTR ratio as a simple indirect method to assess VA deficiency during inflammation. The results show that VA deficiency results in a differential reduction of RBP relative to TTR, which thereby reduces the molar ratio of RBP:TTR. Inflammation, in contrast, reduced both RBP and TTR, but there was no significant differential reduction in RBP and, therefore, no change in the RBP:TTR ratio. Plasma RBP was lowest in marginally VA-deficient rats without inflammation, whereas their plasma TTR concentration was barely affected (Table 1). It is of interest in this regard that Peterson et al. (1974)
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
). The importance of VA in resistance to infectious diseases has been recognized for years, and the association between VA deficiency and increased child mortality has recently been confirmed (Beaton et al. 1994, Sommer and West 1996). As a result, VA supplementation is recommended to reduce the prevalence of VA deficiency and improve child survival. However, implementation of this recommendation requires knowledge of the VA status of the various populations in question.
, Burri et al. 1992). The homeostasis of this complex depends on the adequacy of liver VA stores and on the synthesis of RBP. During the onset of VA deficiency, the hepatic synthesis of RBP continues (Soprano et al. 1982), but the secretion of holo-RBP is markedly diminished (Soprano and Blaner 1994). Therefore, plasma retinol and RBP concentrations decline progressively and nearly in parallel. Numerous studies have demonstrated that the provision of retinol leads to the rapid intrahepatic formation of holo-RBP and its release into plasma (Goodman 1984, Soprano and Blaner 1994). This rapid secretion of holo-RBP is the basis of the relative dose-response (RDR) and the modified relative dose-response tests that have been developed as indicators of hepatic VA reserves (VA status) (Duitsman et al. 1995, Loerch et al. 1979). On the other hand, the induction of inflammation reduces the hepatic synthesis of RBP (Aldred and Schreiber 1993).
). Because of the reduced hepatic synthesis of RBP during inflammation and the reduced release of holo-RBP into plasma, indicators of VA status that are based on measuring the concentration of plasma retinol and/or RBP may be rendered unreliable during infection or inflammation (Rosales and Ross 1998). In this regard, it has been concluded from several field studies that serum retinol is reduced during various infections, but that this reduction does not reflect the actual hepatic stores of VA (Filteau et al. 1993, Samba et al. 1990, Thurnham and Singkamani 1991).
, Rosales et al. 1996b). Second, although TTR, like RBP, is a negative acute-phase reactant (Aldred and Schreiber 1993), the plasma concentration of TTR, in contrast to RBP, is negligibly affected by VA deficiency (Peterson et al. 1974). Therefore, although inflammation may induce reductions in both plasma RBP and TTR concentrations, inflammation per se should have little effect on the RBP:TTR molar ratio. Third, RBP, in contrast to TTR, is expected to be differentially reduced during VA deficiency. If so, then a low ratio of plasma RBP relative to TTR may be a selective indicator of VA deficiency during inflammation. We first tested this hypothesis in a rat model of endotoxin-induced inflammation and then further tested it by analyzing the concentrations of serum RBP and TTR, in relationship to serum retinol and inflammation status, using data from a previously reported randomized, controlled study of Zambian children with acute measles who received either a VA supplement or placebo (Rosales et al. 1996a).
EXPERIMENTAL METHODS
Abstract
Introduction
Methods
Results
Discussion
References
). Briefly, pathogen-free Sprague-Dawley rats (Charles River Breeding Laboratories, Kingston, NY) were housed in a room maintained at 22°C with a 12-h light:dark cycle and were given free access to water and food (Purina rodent chow, Ralston Purina, St. Louis, MO). To induce marginal VA deficiency, a modification of a previously reported protocol (Gardner and Ross 1993) was used in which lactating Sprague-Dawley rats with 4-d-old pups, purchased from Charles River Breeding Laboratories (Wilmington, MA), were fed a purified rodent diet [AIN-93G (Reeves et al. 1993) modified to contain no VA]. This diet was fed during lactation and until pups were 35 d old. For the next 30 d, young rats were fed the same diet containing 0.3 mg retinol (as retinyl palmitate)/kg diet. Marginal VA status was confirmed before experimentation by a reduction of ~50% in plasma retinol compared with VA-sufficient rats (Rosales and Ross 1998).
), or VA-marginal rats of the same age and body weight (Rosales and Ross 1998). In each experiment, food was withdrawn immediately and rats were monitored for signs of inflammation such as a rise in core body temperature >1.0°C. Six hours after LPS or saline, when the core body temperature of LPS-treated rats had risen by 
1.0°C, subgroups of LPS- or saline-treated VA-marginal rats were given a single oral dose of 7.1 µmol of VA (as Aquasol, Astra, Westborough, MA); these groups are referred to as VA-supplemented. The purpose of including this experimental group was to assess the response of plasma RBP and TTR to VA supplementation of marginally VA-deficient rats during inflammation (i.e., similar to RDR test). In this manner, the molar ratio of RBP:TTR was examined under three metabolically different stages of VA status (e.g., VA deficiency vs. VA sufficiency vs. acute VA supplementation). All rats were killed 24 h after LPS administration by CO2 asphyxiation. Approximately 5 mL of blood was obtained from the inferior vena cava in heparinized syringes. The livers were excised, blotted, immediately frozen in liquid nitrogen and stored at 
70°C until they could be processed (Furr et al. 1994).
). The concentrations of RBP and TTR in plasma were determined by sensitive and specific RIA, as described previously (Rosales et al. 1996b, Smith et al. 1975).
). These data had been collected in a VA supplementation trial of children with acute measles conducted in Ndola, Zambia. Materials, methods and procedures for recruitment, enrollment and follow-up of measles patients have been published (Rosales and Kjolhede 1994, Rosales et al. 1996a). Briefly, a total of 200 acute measles patients, aged 5 mo to 17 y, who did not require hospitalization, were enrolled in a randomized, double-masked, placebo-controlled clinical trial. At enrollment, a blood sample was drawn for baseline evaluation, and children with acute measles were allocated, using a one-to-one randomization scheme, to receive a single 210 µmol oral dose of VA or a placebo. Patients were followed for 1 mo with four weekly visits. A second blood sample was obtained at the wk 2 evaluation visit. Blood samples were collected in nonheparinized vacutainers, stored immediately in a cold box at 8°C and allowed to coagulate. Within hours, serum was separated at the Tropical Disease Research Centre and aliquots were stored at 20°C for retinol determination by HPLC using the method of Bieri et al. (1979). Serum proteins were measured by radial immunodiffusion (Mancini et al. 1965) with commercially available kits for TTR, CRP (Kent Laboratories, Redmond, WA) and RBP (Behring Diagnostics, Somerville, NJ). Weight-for-age Z-score distribution was calculated using Center for Disease Control and Prevention Anthropometric software (Atlanta, GA).
). For data on children with measles, monotonic transformations of the variables were necessary to approximate normal probability distributions; for serum retinol, the logarithm (base 10) was used, whereas for RBP, TTR and CRP, the square root transformation was used (Emerson 1991). Nonparametric tests (Mann-Whitney U test, Kruskal-Wallis and Wilcoxon-pair tests) were used when necessary (Rosner 1986). A correlation matrix among variables was obtained with the linear regression procedure of the SPSS statistical package (SPSS, Chicago, IL).
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
View this table:
Table 1.
Plasma retinol, retinol-binding protein (RBP), transthyretin (TTR) and the RBP:TTR ratio in rats without inflammation and with inflammation classified by vitamin A (VA) status1
, Rosales et al. 1996a), 200 were enrolled in the VA supplementation study: 90 in the VA group and 110 in the placebo group. The majority of enrolled children were <60 mo of age, 44.5% were males and more than one third suffered from chronic undernutrition (Rosales and Kjolhede 1994, Rosales et al. 1996a). During acute measles, 63% of children in the VA group and 68% of those in the placebo group had pneumonia, and the rest had cough alone. However 2 wk later, the majority of children were asymptomatic: 49% in the VA group and 58% in the placebo group (see Table 2 of Rosales et al. 1996a).
View this table:
Table 2.
Serum retinol concentration and its transport proteins in children with measles before and after
receiving a placebo or 210 µmol/L of VA1
). Two weeks later, when the majority of children were asymptomatic (convalescent measles), serum retinol, RBP and TTR concentrations were significantly increased, whereas serum CRP had decreased. The significant improvement in serum retinol, RBP and TTR was observed even in children receiving a placebo, and there were no significant differences in these measurements between the VA and placebo groups.
View this table:
Table 3.
Correlation matrix for biochemical and anthropometric variables of children with measles before receiving
a placebo or 210 µmol/L of vitamin A1
View this table:
Table 4.
Serum RBP to TTR molar ratio and CRP concentration by plasma retinol concentration in children
with measles 2 wk after oral placebo or VA1
0.3, Fig. 1B). At baseline, children assigned to receive placebo or VA had similar RBP:TTR ratios (P = 0.43, Fig. 1A, and P = 0.83, Fig. 1B, by Mann-Whitney U test). Two weeks after receiving their respective treatments, there was no response in placebo-treated children with baseline RBP:TTR ratio <0.3 (P = 0.51, Fig. 1A, by Wilcoxon matched-pair test). However, the RBP:TTR ratio of VA-supplemented children had increased significantly, and >60% of VA-treated children had an RBP:TTR ratio >0.3 (P = 0.005, Fig. 1A, by Wilcoxon matched-pair test). In contrast, 2 wk after children with baseline RBP:TTR ratios 0.3 received their respective treatments, their ratios had not changed significantly (P = 0.10 for placebo-treated children, and P = 0.09 for VA-supplemented children, Fig. 1B, by Wilcoxon matched-pair test) and remained well above 0.3.
View larger version (39K):
[in a new window]
Fig 1.
The response of the molar ratio of retinol-binding protein to transthyretin (RBP:TTR) after a placebo or an oral dose of 210 µmol vitamin A (VA) in children with subclinical VA deficiency (i.e., serum retinol < 0.70 µmol/L). Panel A: the distribution of the RBP:TTR ratios among children with baseline ratios < 0.30 during acute measles before treatment (29 children in the placebo and 23 in the VA groups), and during convalescence 2 wk after treatment (n = 22 in the placebo and 20 in the VA groups). Only the group of children who received VA had a significant increase in RBP:TTR ratio. Panel B: the distribution of RBP:TTR ratio among children with baseline ratios 0.30 before treatment (n = 59 in the placebo and 47 in the VA groups), and 2 wk after treatment (n = 36 in the placebo and 41 in the VA groups). *P = 0.005 for post- vs. pretreatment, by Wilcoxon matched pair test.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
previously documented the progressive reduction of serum RBP and TTR concentrations during the course of VA depletion in rats, but these authors did not report the concomitant change in the RBP:TTR molar ratio. Using their data, we calculated this ratio, which is plotted in Figure 2 along with the absolute concentrations previously reported by Peterson et al. (1974). Serum RBP was reduced to 20% of initial values after rats consumed a VA-free diet for 7 wk (Fig. 2A). Serum TTR also decreased, but only to ~75% of initial levels (Fig. 2A). The greater, differential reduction of RBP compared wiith TTR is underscored by the reduction in the RBP:TTR ratio, which declined progressively with increasing severity of VA deficiency, and proportionately with the reduction of serum RBP concentration (Fig. 2B). As was shown in Table 1, this differential reduction in RBP compared with TTR is enhanced by inflammation. Figure 2B superimposes the plasma RBP:TTR ratios from the marginally VA-deficient rats in our study (Table 1) on the curve for the serum RBP:TTR molar ratio calculated from the report of Peterson et al. (1974). The RBP:TTR ratio of marginally VA-deficient rats without inflammation approximates the RBP:TTR ratio of early VA deficiency in the study by Peterson et al. (i.e., in rats consuming VA-free diet for 5 wk). However, the ratio of marginally VA-deficient rats with inflammation corresponds to that of rats with severe VA deficiency (i.e., in rats consuming the VA-free diet for 7 wk). This shows that the biological significance of inflammation is to enhance the differential reduction of RBP to TTR that occurs during VA deficiency. Because of this, the performance of the RBP:TTR ratio to assess VA deficiency is not impaired during inflammation. On the other hand, as discussed below, the determination of RBP (or retinol) concentration alone does not discriminate between low values due to VA deficiency and low values due to infection/inflammation. Indeed, as shown previously (Rosales et al. 1996b), retinol and RBP concentrations were reduced even in VA-sufficient animals during acute inflammation.
View larger version (27K):
[in a new window]
Fig 2.
Concentrations of serum RBP and TTR in rats developing VA deficiency, and their respective RBP:TTR molar ratios. Values shown in panel A were obtained from data of Peterson et al. (1974) and are the means for five rats per group. Panel A: serum RBP and TTR throughout 10 wk of consuming a VA-free diet. Panel B: the RBP:TTR molar ratios calculated from the data of Peterson et al. (1974) in panel A. The mean ± SEM of the RBP:TTR ratios of marginally VA-deficient rats from this study are superimposed.
)]; this value is low even compared with a reference population of 1- to 5-y-old children [i.e., the 2.5 percentile of RBP = 0.48 µmol/L and the 97.5 percentile = 3.61 µmol/L (Lockitch et al. 1988)]. Serum TTR concentration was also very low in these children in comparison to a normal healthy population of 1- to 5-y-olds [i.e., the 2.5 percentile of TTR = 2.5 µmol/L, and the 97.5 percentile = 5.4 µmol/L (Lockitch et al. 1988)]. Serum CRP concentration was increased in acute measles. A significant reduction in the RBP:TTR molar ratio in placebo-treated children with very low serum retinol and severe inflammation is shown in Table 4. This analysis implies that the RBP:TTR ratio performs similarly in humans to what we observed in experimental animals. To assess the internal validity of the RBP:TTR ratio to measure VA deficiency, we selected acute measles patients with baseline serum retinol concentrations <0.7 µmol/L (i.e., subclinical VA deficiency), and RBP:TTR ratios <0.30 (Fig. 1A). This preliminary cut-off value for the RBP:TTR ratio was chosen because it corresponds to the mean value observed in placebo-treated children (those most likely to be VA deficient) with very low serum retinol and severe inflammation (Table 4). Two weeks later, there was a significant increase in the RBP:TTR ratio only in those children who had received VA (Fig. 1A). In contrast, there was no significant change in RBP:TTR ratio 2 wk later in placebo- or VA-treated children whose baseline ratios were 0.30, even if their baseline serum retinol was <0.7 µmol/L. This suggests that such children were not depleted of VA; rather, they could not adequately mobilize retinol as retinol-RBP-TTR during acute infection (Rosales and Ross 1998).
on a population of rural children with clinical and biochemical evidence of severe VA deficiency. The median and lower to upper quartiles of the RBP:TTR molar ratio calculated from their study are 0.16 and 0.10-0.18, respectively. Moreover, the RBP:TTR ratio calculated from data published by the same researchers (Wolde-Gebriel et al. 1993b) on a population with relatively less severe VA deficiency gives median and lower to upper quartiles values of 0.25 and 0.22-0.28, respectively. These results, taken together with those above, provide evidence that acute measles patients having a baseline serum retinol <0.70 µmol/L and an RBP:TTR ratio <0.3 suffered from subclinical VA deficiency. In contrast, using serum retinol concentration alone at baseline, or its increase after VA supplementation, did not provide a meaningful interpretation of the VA status of the children in Table 2, nor did it demonstrate the efficacy of VA supplementation.
). In a similar manner, serum retinol, RBP and TTR concentrations responded in parallel to interleukin-6 (IL-6), a proinflammatory cytokine (Castell et al. 1989). In patients with Plasmodium falciparum infection, the increase of parasitemia was directly proportional to serum IL-6 concentration (Tabone et al. 1992). The correlation coefficient, r, between serum IL-6 and retinol was 0.48; between IL-6 and RBP, 0.45; and between IL-6 and TTR, 0.54 (Tabone et al. 1992).
| |
FOOTNOTES |
|---|
Manuscript received 27 March 1998. Initial reviews completed 30 April 1998. Revision accepted 8 June 1998.
| |
LITERATURE CITED |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  S. H. Gieng and F. J. Rosales Plasma {alpha}1-Acid Glycoprotein Can Be Used to Adjust Inflammation-Induced Hyporetinolemia in Vitamin A-Sufficient, but Not Vitamin A-Deficient or -Supplemented Rats J. Nutr., July 1, 2006; 136(7): 1904 - 1909. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Baeten, M. H. Wener, D. D. Bankson, L. Lavreys, B. A. Richardson, K. Mandaliya, J. J. Bwayo, and R. S. McClelland HIV-1 Infection Alters the Retinol-Binding Protein:Transthyretin Ratio Even in the Absence of the Acute Phase Response J. Nutr., June 1, 2006; 136(6): 1624 - 1629. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Spears, C. Cheney, and J. Zerzan Low plasma retinol concentrations increase the risk of developing bronchopulmonary dysplasia and long-term respiratory disability in very-low-birth-weight infants Am. J. Clinical Nutrition, December 1, 2004; 80(6): 1589 - 1594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Tanumihardjo Assessing Vitamin A Status: Past, Present and Future J. Nutr., January 1, 2004; 134(1): 290S - 293. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. B Stephensen, L. M. Franchi, H. Hernandez, M. Campos, A. Colarossi, R. H Gilman, and J. O Alvarez Assessment of vitamin A status with the relative-dose-response test in Peruvian children recovering from pneumonia Am. J. Clinical Nutrition, December 1, 2002; 76(6): 1351 - 1357. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Rosales, K. K. Chau, M. H. Haskell, and A. H. Shankar Determination of a Cut-Off Value for the Molar Ratio of Retinol-Binding Protein to Transthyretin (RBP:TTR) in Bangladeshi Patients with Low Hepatic Vitamin A Stores J. Nutr., December 1, 2002; 132(12): 3687 - 3692. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Rosales Vitamin A Supplementation of Vitamin A Deficient Measles Patients Lowers the Risk of Measles-Related Pneumonia in Zambian Children J. Nutr., December 1, 2002; 132(12): 3700 - 3703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. P. Erlinger, E. Guallar, E. R. Miller III, R. Stolzenberg-Solomon, and L. J. Appel Relationship Between Systemic Markers of Inflammation and Serum {beta}-Carotene Levels Arch Intern Med, August 13, 2001; 161(15): 1903 - 1908. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
