- Email: [email protected]
Relationship of holo-free and transthyretin-bound plasma retinol-binding protein levels with liver vitamin A in rats
Relationship of holo-free and transthyretin-bound plasma retinol-binding protein levels with liver vitamin A in rats Betty J. Burri, Mark A. Kutnink, and Terry R. Neidlinger B i o c h e m i s t r y Unit, W e s t e r n H u m a n N u t r i t i o n R e s e a r c h Center, U S D A , A R S , Presidio o f San Francisco, CA USA
Vitamin A containing (holo) free and transthyretin-bound (TTR) retinol binding protein (RBP) concentrations in plasma from rats with variable vitamin A status were determined by high performance liquid chromatography. Two different methods were used: (1) molecular exclusion with a TSK 2000 column and (2) reverse phase using a Protesil Octyl 300 column. Holo TTR-RBP peak areas were positively correlated to liver vitamin A concentrations (r = 0.79 for molecular exclusion, 0.81 for reverse phase) in rats with marginal and normal vitamin A status. This correlation was higher than the correlation o f serum retinol to liver vitamin A in these rats (r = 0.58). The correlation o f holoTTR-RBP to liver vitamin A was also higher than its correlations to plasma vitamin A. Therefore, plasma concentrations o f holo-TTR-RBP may be influenced by marginal vitamin A liver stores to a greater extent than plasma retinol is. This suggests that holo-TTR-RBP protein concentrations may be the better indicator o f marginal vitamin A nutritional status in rats. The correlations o f both holo-TTR-RBP and serum retinol decreased sharply at high liver vitamin A concentrations. Neither method is suitable for measuring sub-toxicity in rats.
Keywords: retinol-binding protein; vitamin A; rats; liver Introduction Retinol binding protein (RBP) is the major retinoltransporting protein in blood serum. L2 Usually RBP circulates in blood in a 1:1 c o m p l e x with transthyretin, (TTR) but a significant fraction of free R B P (not c o m p l e x e d to transthyretin) is present in the blood. Immunologically active (apo + holo) R B P concentration has been used to detect vitamin A deficiencies in animals and m a n b e c a u s e it is influenced by the a m o u n t of vitamin A stored in the liver. 36 Unfortunately, other factors such as protein-calorie malnutrition also influence apo + holo RBP, making it unreliable as a m a s s screening test for vitamin A deficiency. 7-12 Holo-free R B P is reportedly a good indi-
A report of preliminary work was presented at the FASEB national meeting in Las Vegas, NV, May 1-6, 1988. Address reprint requests to Betty Jane Burri, Ph.D. USDA/ARS/ WHNRC, P.O. BOX 29997, Presidio of San Francisco, CA, USA 94129. Received March 13, 1991; accepted July 8, 1991. ©
1992 Butterworth-Heinemann
cator o f vitamin A status that is not influenced by alcoholism. ~3 T h a t study did not actually measure holofree RBP, but calculated it from the relative amounts of R B P and T T R in the serum. We have developed two independent high p e r f o r m a n c e liquid c h r o m o t o g r a p h y ( H P L C ) m e t h o d s that quantitate holo-free and holoT T R - R B P separately. We used these techniques to investigate w h e t h e r holo-free, h o I o - T T R - R B P concentrations, or p l a s m a vitamin A concentration is a better indicator of liver vitamin A status. We find that plasma h o l o - T T R - R B P concentrations are a good indicator of vitamin A status in marginally deficient and normal rats, with a high positive correlation to liver vitamin A concentrations. This correlation is higher than the correlation of p l a s m a retinol to liver vitamin A or to apo + holo-RBP.
Materials and methods Instrumentation
Holo-free and holo-TTR-RBP concentrations were assayed on a Model 1084B liquid chromatograph (Hewlett Packard, Avon-
J. Nutr. Biochem., 1992, vol. 3, January
31
Research Communications dale, PA, USA) equipped with an autosampler. Detection was by a Kratos FS 970 fluorometer (ABI Analytical, Ramsey, N J, USA) using a 330 n m excitation wavelength and a 418 nm emission filter. Molecular exclusion H P L C was done on a 7.5 mm x 30 cm Altex Spherogel T S K 2000 analytical column with a 7.5 mm x 7.5 cm Altex Spherogel T S K guard column (Beckman Inst., Fullerton, CA, USA). Reverse phase H P L C was done with a 4.5 m m x 25 cm W h a t m a n Protesil Octyl 300 column, equipped with a 4.5 m m x 25 cm W h a t m a n Solvecon precolumn (Whatman, Clifton, NJ, USA) and a 3.2 mm z 1.5 cm Newguard column (Pierce, Rockford, IL, USA). Fluorescent and UV peak areas were compared on a Perkin Elmer Series 400 chromatograph equipped with two LCI 100 integrators, one for fluorescent detection and the other for UV detection with an LC 90 UV Vis spectrophotometric detector. Serum and liver vitamin A were assayed on a Series 400 liquid chromatograph equipped with an ISS-100 autosampler and an LCI 100 computing integrator (Perkin Elmer, San Jose, CA, USA). Detection was by a Perkin Elmer LC 90 UV-Vis Variable Wavelength S p e c t r o p h o t o m e t e r at 330 nm. Reverse phase H P L C was done with a 4.5 mm x 8.0 cm Perkin Elmer C18 column.
Chemicals and reagents Vitamin A deficient rat diet was from US Biochemical Cleveland, OH, USA; diet 23345). Diet composition is shown in Table 1. Normal rat chow was from Ralston Purina (Richmond, IN, USA; Basal diet 5755). The vitamin A-deficient diet was nutritionally adequate in all nutrients, except vitamin A. Vitamin A concentration in 20 g aliquots of diet was extracted and chromatographed as described ~4~5 with the exception that the extraction procedure was scaled up by a factor of 20. No measurable vitamin A was detected. K e t a m i n e hydrochloride (Vetalar) was from Parke-Davis (Morris Plains, N J, USA). H u m a n RBP standards and purified TTR (prealbumin) were from Behring Diagnostics (San Diego, CA, USA) Retinyl acetate was from Sigma (St. Louis, MO, USA). Purity, as determined by HPLC, was 98%. All other reagents and chemicals were H P L C or reagent grade.
Animals In the first of two experiments, 23 weanling male Sprague Dawley rats (initial weight about 70 g) were housed separately in hanging wire cages in a temperature- and humidity-controlled room with normal 12-hr light : dark cycles. Food and water were available ad libitum throughout the study. Rat weights and food c o n s u m p t i o n (adjusted for spillage) were measured twice a week throughout the study. Rats were fed vitamin A deficient diet for 6 weeks, then divided into four groups (of 5, 6, 6, and 6 animals,
respectively), so that the average weight of each group was similar. Rats in each group were then fed vitamin A deficient diet supplemented with the addition of either 0.5, 1.0, 3.0, or 10.0 mg retinyl palmitate/kg diet for 2 weeks. (The National Research Council r e c o m m e n d s a minimal daily allowance of 4000 IU, or 2.2 mg retinyl palmitate/kg d i e t ) 6 Rats were anesthetized with ketamine hydrochloride, then decapitated. Rat livers were excised, weighed, then quick frozen in liquid nitrogen. Trunk blood was collected ( E D T A anticoagulant), then centrifuged at 2000g for 15 rain at 4 ° C. The livers, red blood cells, and serum were stored at - 2 0 ° C in the dark until use. In the second experiment, male weanling rats were fed rat chow supplemented with 10 mg retinyl palmitate/kg diet for 72 days. All other procedures were as described for the first experiment. All procedures using rats conformed to the National Research Council's guides for care and use of laboratory animals and were approved by the animal use committees of the USDA and of the L e t t e r m a n Army Institute of Research.
Humans Blood samples from h u m a n s with suspected vitamin A deficiencies and toxicities were kindly donated by Dr. Robert Russell, Tufts University, Boston, MA, USA; Harold Zwick, Letterman Army Institute of Research, San Francisco, CA, USA; and Daniel Bankson, Fred Hutchinson Cancer Research Center, Seattle, WA, USA.
H P L C o f retinol esters f r o m liver and serum Plasma and liver retinol and retinol esters were extracted and chromatographed by previously described methods. ~4'j5
Molecular exclusion H P L C o f retinol binding protein Plasma was diluted 1:3 with 0.15 i saline filtered through a 0.20 p. Acro LC 13 filter (Gelman Security, Ann Arbor, M1, USA) and injected directly onto the TSK column. Isocratic elution of 50 ~ L diluted serum was with 0.15 i sodium phosphate, 0.001 M EDTA, 0.002 i [3-mercaptoethanol, pH 7.0 at a flow rate of 0.7 mL/min, t7 Peak identities were confirmed by comparison of retention times to purified TTR, h o l o - f r e e RBP, and molecular weight standards, by extraction of retinol from H P L C peaks, by repeated c h r o m a t o g r a p h y on the reverse phase column, and by comparison to h u m a n RBP.J7 An aliquot from a human plasma pool of known RBP concentration was chromatographed after every 10 rats to correct for possible changes in the chromatographic column.
Table 1 Vitamin A-deficient rat diet Component Dried yeast (vitamin D) Corn starch Vegetable oil Vitamin free casein Vitamin D2 Salt mixture #2 Salt mixture #2 Aluminum potassium sulfate Calcium diphosphate, 2H 2 Potassium diphosphate
32
J. Nutr. B i o c h e m . , 1992, vol. 3, J a n u a r y
g/100 g Diet 8 65 5 18 224 units 4 g/100 g supplement 0.021 9.812 33.309
g/100 g Supplement Calcium carbonate Cobalt chloride, 6H20 Cupric sulfate Ferric citrate, 5 H 2 Magnesium sulfate Manganese sulfate Potassium iodide Sodium chloride Sodium fluoride Sodium tetraborate Zinc chloride
30.985 0.260 0.021 2.840 5.112 0.413 0.083 17.300 0.026 0.026 0.026
Effect of vitamin A on RBP in rats: Burri et al.
Reverse phase H P L C o f retinol binding protein Rat plasma was diluted 1:3 with 13.3% isopropanol and 0.133% trifluoroacetic acid (TFA) in distilled water, vortexed for about 10 sec, then centrifuged at 4000g for 10 min. The precipitate was discarded and the supernatant filtered through a 0.20 I~ Acro LC 13 filter. A 20 ixL aliquot was injected onto a Protesil 300 Octyl Reverse Phase Column (Whatman, Inc., Clifton, N J, USA) to quantify peak 1 (holo-TTR-RBP).To quantify peak 2b (vitamin A derived from holo-free RBP and from an unidentified high molecular weight lipoprotein associated peak), 120 I~L of the diluted sample preparation was reinjected. Elution of both peaks was with a gradient of buffer A (0.1% TFA in distilled water) and buffer B (0.1% TFA in isopropanol). The gradient was started with 10% buffer B for ! min, then increased linearly to 70% buffer B over 19 min. The gradient stabilized at 70% buffer B for 1 min, then decreased linearly to 10% buffer B for 9 min. The gradient was maintained at 10% buffer B over the last 10 rain of the run (total run time 40 min). The flow rate was a constant 1.0 mL/min, with pH constant at 2.2. The identity of peak 1 was confirmed by repeated chromatography on the TSK molecular exclusion column, by extraction of retinol from the HPLC peaks, and by comparison with human RBP. An aliquot from a human plasma pool of known RBP concentration was run after approximately every tenth rat sample, to correct for possible changes in the column.
Comparison o f TSK and reverse phase H P L C The areas of peak 1 (holo-TTR-RBP) and peak 2a (holo-free RBP) from the TSK 2000 molecular exclusion column were correlated to the areas of peak 1 and peak 2b derived from the reverse phase Protesil Octyl 300 column. Linear correlation coefficients were derived from SAS (Statistical Analysis System, Cary, NC, USA), using the best least squares fit lines.
Quantitation o f holo-RBP Holo-free and holo-TTR-RBP concentrations were estimated by three previously described ~7 independent methods. These methods were: (1) extraction and quantitation of retinol from the HPLC peaks; (2) comparison to purified human vitamin Asaturated free RBP; and (3) comparison of UV detection at 330 nm with fluorescent detection at 330 nm excitation, 418 emission. The estimates from each method were averaged to get the reported RBP concentration.
Estimation o f apo R B P concentrations A small amount of a concentrated solution of retinol in 95% ethanol was added to samples of rat plasma so that the retinol was present in a 4 to 10 times excess molar concentration relative to RBP. These vitamin A-saturated preparations were incubated overnight at 4° C, then run on the TSK HPLC column using standard conditions. Apo RBP concentrations were estimated by subtracting the peak areas of rat plasma incubated with 95% ethanol alone from the peak areas of rat plasma incubated with retinol. Peak areas for the unknown samples were compared to HPLC peak areas of known standards. First order linear regression coefficients were derived with Sigmaplot (Jandel Scientific, Sausalito, CA, USA) and with SAS and were based on the best least squares fit of the data.
Results T h e rats u s e d in t h e first s t u d y h a d v i t a m i n A l i v e r s t o r e s r a n g i n g f r o m m a r g i n a l l y d e f i c i e n t to a d e q u a t e . T h e r a t s in t h e s e c o n d s t u d y w e r e a d e q u a t e o r sub-
toxic. Most rats had adequate liver vitamin A stores at t h e e n d o f t h e e x p e r i m e n t s . A t no t i m e d i d a n y o f t h e s e r a t s s h o w s y m p t o m s o f g r o s s v i t a m i n A defic i e n c y o r t o x i c i t y . A l l r a t s c o n t i n u e d to gain w e i g h t during the study, though the rate of growth was s l o w e d d u r i n g t h e l a s t w e e k t h e r a t s w e r e f e d the uns u p p l e m e n t e d v i t a m i n A - d e f i c i e n t diet. C o a t s , e y e s , and other physical symptoms remained normal. Rat w e i g h t s , l i v e r w e i g h t s , a n d h e m a t o c r i t s w e r e n o t sign i f i c a n t l y d i f f e r e n t in a n y g r o u p o f rats at t h e b e g i n n i n g or end of the times they were on different vitamin A supplements. Liver and plasma vitamin A concentrat i o n s , h o w e v e r , d i d d i f f e r a m o n g g r o u p s . T h e rats g i v e n t h e l o w e s t v i t a m i n A s u p p l e m e n t h a d the l o w e s t l i v e r v i t a m i n A c o n c e n t r a t i o n s (Table 2). R a t h o l o - R B P h a s v e r y s i m i l a r p r o p e r t i e s to h u m a n h o l o - R B P o n m o l e c u l a r e x c l u s i o n H P L C 17 (Figures la and Ib). T h r e e p e a k s a r e p r e s e n t in m o l e c u l a r exc l u s i o n c h r o m a t o g r a m s o f R B P . T h e 17-minute p e a k ( p e a k 1) h a s b e e n i d e n t i f i e d as h o l o - T T R - R B P . 17 T h e 2 4 - m i n u t e p e a k ( p e a k 2a) h a s b e e n identified as h o i o free RBP.17 T h e t h i r d p e a k is i d e n t i f i a b l e in all o f o u r c h r o m a t o g r a m s f r o m rat p l a s m a b u t o n l y s o m e o f o u r chromatograms from human plasma, This peak (peak 0) m i g r a t e s at 12 m i n u t e s a n d a p p e a r s to be a high m o l e c u l a r w e i g h t c o m p l e x that c o n t a i n s v i t a m i n A a n d lipoprotein. T h e s i m i l a r i t y o f rat a n d h u m a n h o l o - R B P on rev e r s e p h a s e c h r o m a t o g r a p h y is e q u a l l y striking, as s h o w n b y c o m p a r i s o n s o f Figures 2a and 2b. T w o p e a k s a r e p r e s e n t in the r e v e r s e d p h a s e c h r o m a t o g r a m o f e i t h e r rat o r h u m a n s e r u m , a n d b o t h a r e in s i m i l a r l o c a t i o n s . T h e m a j o r p e a k , ( p e a k 1), r e a c t s with antib o d i e s to R B P a n d to T T R in h u m a n s e r u m a n d has b e e n i d e n t i f i e d as h o l o - T T R - R B P b e c a u s e o f its f l u o r e s c e n t p r o p e r t i e s , its p h y s i o c h e m i c a l p r o p e r t i e s , a n d its v i t a m i n A c o n t e n t (Table 3). T h e small p e a k ( p e a k 2b) h a s b e e n i d e n t i f i e d as v i t a m i n A d e r i v e d f r o m h o l o - f r e e R B P a n d f r o m the high m o l e c u l a r w e i g h t l i p o p r o t e i n - a s s o c i a t e d c o m p l e x , b a s e d on its f l u o r e s c e n t a n d p h y s i o c h e m i c a l p r o p e r t i e s , its v i t a m i n A c o n t e n t , a n d its l a c k o f r e a c t i v i t y to e i t h e r R B P o r t r a n s t h y r e t i n a n t i b o d i e s in h u m a n s e r u m . T h e r e usually is o n l y a s m a l l a m o u n t o f l i p o p r o t e i n - a s s o c i a t e d v i t a m i n A in h u m a n p l a s m a , so p e a k 2b c o r r e s p o n d s c l o s e l y to p e a k 2a of Figure la and lb ( f r e e - h o l o R B P o n t h e m o l e c u l a r e x c l u s i o n c o l u m n ) in h u m a n s . P e a k 2b in rats h a s high c o r r e l a t i o n s w i t h the high m o l e c u l a r weight lipoprotein associated vitamin A complex ( p e a k 0, Figure la) a n d a l e s s e r c o r r e l a t i o n to f r e e h o l o - R B P ( p e a k 2a, Figure la), b e c a u s e it c o n s i s t s o f vitamin A derived from both complexes. H o l o - T T R - R B P d o e s n o t d i s s o c i a t e in e i t h e r the m o l e c u l a r e x c l u s i o n 17 o r t h e r e v e r s e p h a s e m e t h o d w h e n t h e h o l o - T T R - R B P p e a k is c o l l e c t e d a n d rechromatographed three times. Molecular exclusion HPLC and reverse phase H P L C p e a k a r e a s a r e h i g h l y c o r r e l a t e d to e a c h o t h e r (Figure 3). T h e p e a k a r e a s o f p e a k 1 (identified as h o l o - T T R - R B P ) s h o w e d t h e h i g h e s t c o r r e l a t i o n s bet w e e n t h e s e t w o m e t h o d s (r = 0.96). T h e p e a k a r e a s o f p e a k 2a ( f r e e - h o l o R B P , m o l e c u l a r e x c l u s i o n ) a n d
J. Nutr. Biochem., 1992, vol. 3, January
33
Research Communications Table 2
Physiochemical properties of rats fed different amounts of vitamin A
Mg RP Kg diet
n
0.5 1.0 3.0 10.0 Study 2
5 6 6 6 6
Rat liver weight (final) g
Rat weight (final) 433 422 418 448 514
-+ 58 _+ 63 + 22 -+ 30 _+ 17
14.2 16.9 16.3 16.9 16.7
Liver vitamin A I~g RE g liver
-+ 2.7 -+ 3.8 _+ 18 -+ 1,3 _+ 0.6
26.8 32.8 29.4 67.5 352
+ 16.3 -+ 7.4 _+ 7.4 -+ 10.7 ~ _+ 149 a
RP, retinyl palmitate; RE, retinol and retinyl esters. aSignificant difference, P < 0.05.
2b (free-holo RBP and lipoprotein associated vitamin A, reverse phase) are not as highly correlated in these two methods (r = 0.64). There is a higher correlation between peaks 0 (lipoprotein associated vitamin A, molecular exclusion) and 2b (reverse phase), but only in the rat. This is because reverse phase chromatography disassociates retinol from both holo-free RBP and from the unidentified high molecular weight complex of lipoprotein associated with vitamin A. Apo-free and a p o - T T R - R B P are not present in significant concentrations in these rats. The difference in fluorescence between plasma incubated with ethanol or with saturating amounts of vitamin A ranged from - 6 % to + 14%, and were not dependent on the amount of vitamin A fed to the rats. The effects of liver vitamin A concentration on holo-free and h o l o - T T R - R B P in the first experiment, as measured by molecular exclusion chromatography, is shown in Figure 4. H o l o - T T R - R B P (peak 1) was
positively correlated to liver vitamin A stores (r = 0.79) in marginal or adequate rats. The high molecular weight complex containing lipoprotein and vitamin A is also positively correlated to liver vitamin A (r = 0.79). Holo-free RBP, however, has no significant correlation to liver vitamin A (r = 0.14). These correlations are similar to those derived independently from reverse phase H P L C (Figure 5). H o l o - T T R - R B P has an almost identical correlation with liver vitamin A (r -- 0.81). The correlation of reverse phase peak 2b (hoio-free RBP and the high molecular weight vitamin
c:
8 _= LI-
1
o¢ 0
5'
' 10
' 15
2= U_
0
i
i
i
i
I
I
6
12
18
24
30
36
' 2'o 25 Time (rain)
' 30
3'5
Figure 2a Typical reverse phase chromatogram of rat holo-RBP. Chromatogram is of 20 ILL 1:3 diluted rat plasma, HPLC method is as described in the text.
Time (min)
Figure l e
Typical molecular exclusion (TSK) chromatogram of rat holo-RBP. Chromatogram is of 25 ~L 1:3 diluted rat plasma. HPLC method is as described in text.
c ffl ®
2b IJ _
LI.
0
6
12
18
2'4
3'0
0
Time (rain) Figure l b
Typical molecular exclusion (TSK) chromatogram of human holo-RBP. Chromatogram is of 25 ~L of 1:3 diluted human plasma. HPLC method is as described in text.
34
J. Nutr. Biochem., 1992, vol. 3, January
I
5
I
10
I
15
2'o
30
35
Time (rain)
Figure 2b
Typical reverse phase chromatogram of human holoRBP. Chromatogram is of 100 ILL 1:3 diluted human plasma. HPLC method is as described in the text.
Effect of vitamin A on RBP in rats. Burri et al.
Table 3 Properties of HPLC peaks from normal rat plasma (peak assignment)
Retention time, min Reverse phase Molecular exclusion Molecular mass, Da % of total peak area Reverse phase Molecular exclusion % of total extractable vitamin A in peak Reverse phase Molecular exclusion Excitation max, nm Molecular exclusion Emission max, nm Molecular exclusion
Lipid-protein
holo-TTR-RBP
halo-free RBP
-12.5 Exclusion
17.1 17.8 68,000
24.8 25.5 20,500
-2-8
89-95 23-28
5-11 69-75
-3-7
84-95 82-91
2-11 3-14
365
340
335
465
450
450
A complex) to liver vitamin A is insignificant (r = 0.21), probably reflecting the major contribution of holo-free RBP to this peak. The correlations of holoTTR-RBP to liver vitamin A are higher than the correlation of plasma vitamin A to liver vitamin A in these rats (Figure 6, r = 0.58). The correlations for holoTTR-RBP and retinol with liver vitamin A were not good in adequate-to-subtoxic rats. The correlations of liver retinyl ester concentrations with plasma retinol, holo-TTR-RBP and holo-free RBP were 0.26, 0.34, and 0.27, respectively. Preliminary data on vitamin A-deficient humans (plasma retinol 0.2 to 0.4 ~mol/L) showed that both holo-TTR-RBP and free-holo RBP were decreased significantly (P -< 0.01). Serum retinyl esters increased significantly (P -< 0.001) in humans with vitamin A toxicity. Serum retinol, holo-TTR-RBP and free-holo RBP did not increase significantly from the normal human range. Typical molecular exclusion chromatograms for human plasma RBP in deficiency and toxicity are shown in Figures 7a and 7b. Discussion
ing molecular exclusion chromatography on a TSK 2000 column, the other reverse phase chromatography on a Protesil Octyl 300 column. These methods give very similar results for the effect of liver and plasma vitamin A concentrations on holo-TTR-RBP. Molecular exclusion chromatography may be the superior method of the two because it does not disassociate vitamin A from halo-free RBP and from the high molecular weight vitamin A-containing complex associated with lipoprotein, and it requires less sample preparation. Neither method is species specific, and both can assay vitamin A-containing halo-free and holoTTR-RBP. The HPLC methods described are not strictly comparable to immunologic methods, because the HPLC methods measure holo-RBP only, while immunologic methods measure apo-RPB and any other protein with cross-reactivity to RBP antibodies. Both the HPLC and immunologic methods should usually give almost the same total RBP concentrations (within experimental error) because apo-RBP is present in very small amounts in plasma and serum. Our results showed differences of - 6 % to 14% between vitamin A-saturated and non-saturated plasmas in these rats,
We developed two independent HPLC methods for measuring holo-free and holo-TTR-RBP, one involv2000-
5
0
=
Transthyretin-bound
•
=
Free RBP (Peak
RBP
(Peak 1)
2a) O
0 P e a k 1 RP vs Peak 1 TSK P e a k 2 b RP vs Peak 20 TSK
• 1500
& P e a k 1 RP vii P e a k 0 TSK "6
3
1000
o.. 1-
r - 0.635 •
~
O~
o
5oo
J
j •
0
~
O 009..1 ~
oO~.o
O
r - 0.923 A . " .. " ~ . . .A ~ . -
o o
. .........
"
r = 0.794
=E
3
~
2 0
1
~'~''&
.....
~A.~'~,
0 0
600
1200
Reverse
1800
Phase
Peak
2400
3000
Areas
Figure 3 Correlation of peak areas determined by molecular exclusion HPLC with peak areas determined by reverse phase. Methods as described in text.
o. • • ~
~ I
0
• .oqk.~.e.
20
om ....
~
....
•
.......
~
40 60 Uver Retinyl Esters (ug/g)
••
BO
Figure 4 Correlation of liver vitamin A concentration with halofree and holo-TrR-RBP concentrations in rat plasma, as measured by molecular exclusion HPLC. Methods are described in the text. Each point represents the mean of duplicate assays. © = peak 1 (holo-TTR-RBP). • = peak 2a (halo-free RBP). J. Nutr. B i o c h e m . ,
1 9 9 2 , v o l . 3, J a n u a r y
35
Research Communications 6. O - Tr~anmthymtin-bound RBP (Peak 1) • - Retlnol f r o m f r e e R B P a n d Lipoprotein--=ssociated Vitamin A (Peak 2b)
5 i
O
4 o ~
0
tz
2
o
1
~
O
O m
0 ~
i
V
~
0
r = 0.213
. = mmlL •
•
mmmrR~
mm
20
~m
m
m
~
40
~
60
--
80
U v e r Retinyl Esters ( u g / g )
Figure 5 Correlation of liver vitamin A concentration with h a l o free- and holo-TTR-RBP concentrations in rat plasma, as measured by reverse phase HPLC. Methods are described in the text Each point represents the mean of duplicate assays. © = peak 1 (holo-TTR-RBP). • = peak 2b (vitamin A from halo-free RBP and prepeak)
were always run with at least one human plasma control, an accurate comparison of peak areas of typical chromatograms of RBP from rats and humans was possible. This comparison shows that the unidentified high molecular weight complex associated with lipoprotein and vitamin A (peak 0, molecular exclusion) and holo-TTR-RBP (peak l) concentrations were consistently higher in rats than in humans with similar plasma vitamin A concentrations. Holo-free-RBP concentrations were lower in rats, particularly at higher plasma vitamin A concentrations. There are two probable explanations for these differences: (1) the formation or degradation of the holo-TTR-RBP complex is regulated by different mechanisms in rats
Deficiency 2500
,
2500
2000
2000
1500
1500
5. o
~-4-
o
E ~3.
j
r=0
~ 2-
o
8
o o o
o
qb o
o o
,
1000 PIs~
~
I
,°°° i
500
500
-500
-500
o
o
o o
o 0
.....
0
, .....
20
, .....
40
, .....
60
, .....
80
100
U v e r Retinyl P a l m i t a t e (ug / g)
Figure 6 Correlation of liver vitamin A concentration with plasma vitamin A concentration. Liver retinot esters and plasma retinal were determined by isocratic HPLC as described in text All assays were done in d u p l i c a t e
with no consistent effect of vitamin A concentration in the diet. This study shows holo-TTR-RBP concentrations were better than plasma retinal concentrations as indicators of liver vitamin A stores in marginally deficient-to-normal rats. The correlations between holo-TTR-RBP and liver vitamin A concentrations were significantly better (P = 0.01) than the correlation of plasma vitamin A concentrations to liver vitamin A concentrations. This suggests that holo-TTRRBP concentrations may be influenced by vitamin A liver stores to a greater extent than plasma vitamin A is in marginally deficient rats, and may be better indicators of vitamin A status in normal rats. Preliminary human data suggests that serum retinal, halo-free RBP and holo-TTR-RBP all decrease significantly in vitamin A deficiency. Data from humans and rats suggests that these relationships break down during vitamin A subtoxicity or toxicity. In toxicity, retinyl esters were increased significantly in humans, but serum retinal, holo-TTRRBP and halo-free RBP concentrations were not. Our rat and human chromatograms always showed characteristic differences. Since rat plasma samples
36
J Nutr Biochem., 1992, vol. 3, January
-1000
10
0
20
30
40-1000
Molecular exclusion chromatogram of human haloRBP in vitamin A deficiency Chromatogram is of 25 IxL of 1:3 diluted plasma• HPLC method is as described in text Figure 7a
Toxicity 2500
2500
i
2000
2000
Plwk t 1500
1500
•
pe.L.
1000
•
|
500 ~
:2 -500
t
-1000
0
1'0
0
-500
70
40-- 1000
(.~.ntJm)
Molecular exclusion chromatogram of human haloRBP in vitamin A toxicity. Chromatogram is of 25 ixL of 1:3 diluted plasma. HPLC method is as described in text. Fibure 7b
Effect of vitamin A on RBP in rats: Burri et al.
and humans; (2) the formation of the holo-TTR-RBP complex is influenced by nutritional factors other than vitamin A status. Little is known about the factors influencing the formation of the holo-TTR-RBP complex in vivo, but there is some support for both possibilities. Rat and human RBP have different amino acid compositions, ~8 and different immunologic reactivities. ~9Because RBP seems to be different in these species, it would not be surprising if the binding of RBP to TTR was different. It is also true that the typical diets of rats and humans are different. Human data suggested that free RBP is affected to a much greater extent by changes in liver function (shown in subjects with fluctuating creatinine) or by daily changes in nutrient intake (as indicated by greater day-to-day variations in the free RBP peak. 2° It is possible that RBP levels in this study may have been influenced by differences in dietary factors such as zinc, 2~ protein or calories 1°-12or carbohydrate, z: The rats used in these studies have liver vitamin A stores which were marginally deficient to subtoxic. This narrow range of vitamin A nutriture was used in an attempt to mimic the normal range of vitamin A nutriture in the human populations of the United States, where gross vitamin A deficiency or toxicity is uncommon, but marginal status occurs fairly often. 23 It appears that holo-TTR-RBP may be a good indicator of vitamin A status in this range of vitamin A nutriture. Unfortunately, it is unclear whether the rat is a good model for humans, since it is possible that the binding of TTR to RBP is regulated by different mechanisms in these species. The mechanisms controlling the formation of the holo-TTR-RBP complex require more study than they have been given. The molecular exclusion HPLC method described should provide a useful tool for studying this complex and the mechanisms controlling its formation and degradation in animals and man.
Acknowledgments We would like to thank Peter Taylor, Pearl Sun, and Denise Gretz for their technical assistance. We would also like to thank Drs. Robert Russell, Daniel Bankson, Harold Zwick, and Mary Kretsch for kindly donating blood samples. References 1 2 3 4 5
Goodman, D.S. (1984). Plasma retinol-binding protein. In The Retinoids. (A.B. Sporn and D.S. Goodman, eds.) vol. 2, pp. 41-88, Academic Press, Orlando, FL Goodman, D.S. (1980). Plasma retinol-binding protein. Ann. N . Y . Acad. Sci. 348, 378-390 Muto, Y., Smith, J.E., Milch, P.O., and Goodman, D.S. (1972). Regulation of retinol-binding protein metabolism by vitamin A status in the rat. J. Biol. Chem. 247, 2542-2550 Mallia, A.K., Smith, J.E., and Goodman, D.S. (1975). Metabolism of retinol-binding protein and vitamin A during hypervitaminosis in the rat. J. Lipid Res. 16, 180-188 Harrison, E.H., Smith, J.E., and Goodman, D.S. (1980). Effects of vitamin A deficiency on the levels and distribution of
6
7 8
9 10
11
12
13 14 15 16 17
18 19 20
21
22
23
retinol-binding protein and marker enzymes in homogenates and Golgi-Rich fraction of rat liver. Biochimica et Biophysica Acta 628, 489-497 Audisio, M., Dante, D., Fidanza, F., Villani, C., and Rulli, G. (1985). Valori plasmatici della vitamina A e dei suoi vettori proteici (rbp e PA), del corotene, del colesterolo total e del colesterolo-HDL in soggetti affetti da carcinoma mammario. Boll. Soc. It. Biol. Sper. 61,467-474 Smith, F.R. and Goodman, D.S. (1971). The effects of diseases of the liver, thyroid, and kidneys on the transport of vitamin A in human plasma. J. Clin. Invest. 50, 2426-2436 Smith, F.R., Suskind, R. Thanangkul, O., Leitzmann, C., Goodman, D.S., and Olson, R.E. (1975). Plasma vitamin A, retinol-binding protein and prealbumin concentrations in protein-calorie malnutrition. A m J. Clin. Nutr. 28, 732-738 Smith, F. R., Underwood, B., Denning C., Varma, A., and Goodman, D.S. (1972). Depressed plasma retinol binding protein levels in cystic fibrosis. J. Lab. Clin. Med. 80, 423-433 Smith, F.R., Goodman, D.S., Zaklama, M.S.. Gabr, M., Maraghy, S.E., and Batwardhan, W. (1973). Serum vitamin A, retinol-binding protein, and prealbumin concentration in protein calorie malnutrition. I. Functional defect in hepatic retinol release. Am. J. Clin. Nutr. 26, 973-981 Smith, F.R., Goodman, D.S., Arroyave, G., and Viteri, F. (1973). Serum vitamin A, retinol-binding protein, and prealbumin concentrations in protein-calorie malnutrition. 11. Treatment including supplemental vitamin A. Am. J. Clin. Nutr. 26, 982-987 lngenbleek, Y., Van den Schrieck, H.G., De Nayer, P., and De Visscher, M. (1975). Albumin, transferrin and the thyroxine-binding prealbumin/retinol-binding protein (TBPARBP) complex in assessment of malnutrition. Clin. Chim. Acta 63, 61-67 Fex, G. and Felding, P. (1984). Factors affecting the concentration of free holo retinol-binding protein in human plasma. Eur. J. Clin. Invest. 14, 146-149 Catignani, G.L. and Bieri, J.G. (1983). Simultaneous determination of retinol and c~-tocopherol in serum or plasma by liquid chromatography. Clin. Chem. 29, 708-712 Chow, F.I. and Omaye, S.K. (1983). Use of antioxidants in the analysis of vitamins A and E in mammalian plasma by high performance liquid chromatography. Lipids. 18, 837-841 National Research Council (1978). Nutrient requirements of laboratory animals, vol. 10, 3rd ed., pp. 7-37. National Research Council, Washington, D.C. Burri, B.J. and Kutnink. M.A. (1989). Liquid-chromatographic assay for free and transthyretin-bound retinolbinding protein in serum from normal humans. Clin. Chem. 35, 582-586 Rask, L., Anundi, H., and Peterson, P.A. (1979). The primary structure of the human retinol-binding protein. FEBS Letters 104, 55-58 Muto, Y., Smith, F.. and Goodman, D.S. (1973). Comparative studies of retinol transport in plasma. J. Lipid Res. 14, 525-532 Burri, B.J., Bankson, D.C., and Neidlinger, T.R. (1990). Use of free and transthyretin-bound retinol-binding protein in serum as tests of vitamin A status in humans: Effect of high creatinine concentrations in serum. Clin. Chem. 36, 674-676 Shingwekar, A.G., Mohanram, M., and Reddy, V. (1979). Effects of zinc supplementation on plasma levels of vitamin A and retinol-binding protein in malnourished children. Clin. Chem. Acta 93, 97-100 Kelleher, P.C., Phinney, S.D., Sims, E.A.H., Bogardus, C., Horton, E.S., Bistrian, B.R., Amatruda, J.M., and Lockwood, D.H. (1983). Effects of carbohydrate-containing and carbohydrate-restricted hypocaloric and eucaloric diets on serum concentrations of retinol-binding protein, thyroxinebinding prealbumin and transferrin. Metabolism 32, 95-101 Pilch, S.M. (1985). Assessment of the vitamin A nutritional Status of the U.S. Population based on data collected in the health and nutrition examination surveys. LSRO, FASEB Bethesda, MD
J. Nutr. Biochem., 1992, vol. 3, January
37
Vitamin A containing (holo) free and transthyretin-bound (TTR) retinol binding protein (RBP) concentrations in plasma from rats with variable vitamin A status were determined by high performance liquid chromatography. Two different methods were used: (1) molecular exclusion with a TSK 2000 column and (2) reverse phase using a Protesil Octyl 300 column. Holo TTR-RBP peak areas were positively correlated to liver vitamin A concentrations (r = 0.79 for molecular exclusion, 0.81 for reverse phase) in rats with marginal and normal vitamin A status. This correlation was higher than the correlation o f serum retinol to liver vitamin A in these rats (r = 0.58). The correlation o f holoTTR-RBP to liver vitamin A was also higher than its correlations to plasma vitamin A. Therefore, plasma concentrations o f holo-TTR-RBP may be influenced by marginal vitamin A liver stores to a greater extent than plasma retinol is. This suggests that holo-TTR-RBP protein concentrations may be the better indicator o f marginal vitamin A nutritional status in rats. The correlations o f both holo-TTR-RBP and serum retinol decreased sharply at high liver vitamin A concentrations. Neither method is suitable for measuring sub-toxicity in rats.
Keywords: retinol-binding protein; vitamin A; rats; liver Introduction Retinol binding protein (RBP) is the major retinoltransporting protein in blood serum. L2 Usually RBP circulates in blood in a 1:1 c o m p l e x with transthyretin, (TTR) but a significant fraction of free R B P (not c o m p l e x e d to transthyretin) is present in the blood. Immunologically active (apo + holo) R B P concentration has been used to detect vitamin A deficiencies in animals and m a n b e c a u s e it is influenced by the a m o u n t of vitamin A stored in the liver. 36 Unfortunately, other factors such as protein-calorie malnutrition also influence apo + holo RBP, making it unreliable as a m a s s screening test for vitamin A deficiency. 7-12 Holo-free R B P is reportedly a good indi-
A report of preliminary work was presented at the FASEB national meeting in Las Vegas, NV, May 1-6, 1988. Address reprint requests to Betty Jane Burri, Ph.D. USDA/ARS/ WHNRC, P.O. BOX 29997, Presidio of San Francisco, CA, USA 94129. Received March 13, 1991; accepted July 8, 1991. ©
1992 Butterworth-Heinemann
cator o f vitamin A status that is not influenced by alcoholism. ~3 T h a t study did not actually measure holofree RBP, but calculated it from the relative amounts of R B P and T T R in the serum. We have developed two independent high p e r f o r m a n c e liquid c h r o m o t o g r a p h y ( H P L C ) m e t h o d s that quantitate holo-free and holoT T R - R B P separately. We used these techniques to investigate w h e t h e r holo-free, h o I o - T T R - R B P concentrations, or p l a s m a vitamin A concentration is a better indicator of liver vitamin A status. We find that plasma h o l o - T T R - R B P concentrations are a good indicator of vitamin A status in marginally deficient and normal rats, with a high positive correlation to liver vitamin A concentrations. This correlation is higher than the correlation of p l a s m a retinol to liver vitamin A or to apo + holo-RBP.
Materials and methods Instrumentation
Holo-free and holo-TTR-RBP concentrations were assayed on a Model 1084B liquid chromatograph (Hewlett Packard, Avon-
J. Nutr. Biochem., 1992, vol. 3, January
31
Research Communications dale, PA, USA) equipped with an autosampler. Detection was by a Kratos FS 970 fluorometer (ABI Analytical, Ramsey, N J, USA) using a 330 n m excitation wavelength and a 418 nm emission filter. Molecular exclusion H P L C was done on a 7.5 mm x 30 cm Altex Spherogel T S K 2000 analytical column with a 7.5 mm x 7.5 cm Altex Spherogel T S K guard column (Beckman Inst., Fullerton, CA, USA). Reverse phase H P L C was done with a 4.5 m m x 25 cm W h a t m a n Protesil Octyl 300 column, equipped with a 4.5 m m x 25 cm W h a t m a n Solvecon precolumn (Whatman, Clifton, NJ, USA) and a 3.2 mm z 1.5 cm Newguard column (Pierce, Rockford, IL, USA). Fluorescent and UV peak areas were compared on a Perkin Elmer Series 400 chromatograph equipped with two LCI 100 integrators, one for fluorescent detection and the other for UV detection with an LC 90 UV Vis spectrophotometric detector. Serum and liver vitamin A were assayed on a Series 400 liquid chromatograph equipped with an ISS-100 autosampler and an LCI 100 computing integrator (Perkin Elmer, San Jose, CA, USA). Detection was by a Perkin Elmer LC 90 UV-Vis Variable Wavelength S p e c t r o p h o t o m e t e r at 330 nm. Reverse phase H P L C was done with a 4.5 mm x 8.0 cm Perkin Elmer C18 column.
Chemicals and reagents Vitamin A deficient rat diet was from US Biochemical Cleveland, OH, USA; diet 23345). Diet composition is shown in Table 1. Normal rat chow was from Ralston Purina (Richmond, IN, USA; Basal diet 5755). The vitamin A-deficient diet was nutritionally adequate in all nutrients, except vitamin A. Vitamin A concentration in 20 g aliquots of diet was extracted and chromatographed as described ~4~5 with the exception that the extraction procedure was scaled up by a factor of 20. No measurable vitamin A was detected. K e t a m i n e hydrochloride (Vetalar) was from Parke-Davis (Morris Plains, N J, USA). H u m a n RBP standards and purified TTR (prealbumin) were from Behring Diagnostics (San Diego, CA, USA) Retinyl acetate was from Sigma (St. Louis, MO, USA). Purity, as determined by HPLC, was 98%. All other reagents and chemicals were H P L C or reagent grade.
Animals In the first of two experiments, 23 weanling male Sprague Dawley rats (initial weight about 70 g) were housed separately in hanging wire cages in a temperature- and humidity-controlled room with normal 12-hr light : dark cycles. Food and water were available ad libitum throughout the study. Rat weights and food c o n s u m p t i o n (adjusted for spillage) were measured twice a week throughout the study. Rats were fed vitamin A deficient diet for 6 weeks, then divided into four groups (of 5, 6, 6, and 6 animals,
respectively), so that the average weight of each group was similar. Rats in each group were then fed vitamin A deficient diet supplemented with the addition of either 0.5, 1.0, 3.0, or 10.0 mg retinyl palmitate/kg diet for 2 weeks. (The National Research Council r e c o m m e n d s a minimal daily allowance of 4000 IU, or 2.2 mg retinyl palmitate/kg d i e t ) 6 Rats were anesthetized with ketamine hydrochloride, then decapitated. Rat livers were excised, weighed, then quick frozen in liquid nitrogen. Trunk blood was collected ( E D T A anticoagulant), then centrifuged at 2000g for 15 rain at 4 ° C. The livers, red blood cells, and serum were stored at - 2 0 ° C in the dark until use. In the second experiment, male weanling rats were fed rat chow supplemented with 10 mg retinyl palmitate/kg diet for 72 days. All other procedures were as described for the first experiment. All procedures using rats conformed to the National Research Council's guides for care and use of laboratory animals and were approved by the animal use committees of the USDA and of the L e t t e r m a n Army Institute of Research.
Humans Blood samples from h u m a n s with suspected vitamin A deficiencies and toxicities were kindly donated by Dr. Robert Russell, Tufts University, Boston, MA, USA; Harold Zwick, Letterman Army Institute of Research, San Francisco, CA, USA; and Daniel Bankson, Fred Hutchinson Cancer Research Center, Seattle, WA, USA.
H P L C o f retinol esters f r o m liver and serum Plasma and liver retinol and retinol esters were extracted and chromatographed by previously described methods. ~4'j5
Molecular exclusion H P L C o f retinol binding protein Plasma was diluted 1:3 with 0.15 i saline filtered through a 0.20 p. Acro LC 13 filter (Gelman Security, Ann Arbor, M1, USA) and injected directly onto the TSK column. Isocratic elution of 50 ~ L diluted serum was with 0.15 i sodium phosphate, 0.001 M EDTA, 0.002 i [3-mercaptoethanol, pH 7.0 at a flow rate of 0.7 mL/min, t7 Peak identities were confirmed by comparison of retention times to purified TTR, h o l o - f r e e RBP, and molecular weight standards, by extraction of retinol from H P L C peaks, by repeated c h r o m a t o g r a p h y on the reverse phase column, and by comparison to h u m a n RBP.J7 An aliquot from a human plasma pool of known RBP concentration was chromatographed after every 10 rats to correct for possible changes in the chromatographic column.
Table 1 Vitamin A-deficient rat diet Component Dried yeast (vitamin D) Corn starch Vegetable oil Vitamin free casein Vitamin D2 Salt mixture #2 Salt mixture #2 Aluminum potassium sulfate Calcium diphosphate, 2H 2 Potassium diphosphate
32
J. Nutr. B i o c h e m . , 1992, vol. 3, J a n u a r y
g/100 g Diet 8 65 5 18 224 units 4 g/100 g supplement 0.021 9.812 33.309
g/100 g Supplement Calcium carbonate Cobalt chloride, 6H20 Cupric sulfate Ferric citrate, 5 H 2 Magnesium sulfate Manganese sulfate Potassium iodide Sodium chloride Sodium fluoride Sodium tetraborate Zinc chloride
30.985 0.260 0.021 2.840 5.112 0.413 0.083 17.300 0.026 0.026 0.026
Effect of vitamin A on RBP in rats: Burri et al.
Reverse phase H P L C o f retinol binding protein Rat plasma was diluted 1:3 with 13.3% isopropanol and 0.133% trifluoroacetic acid (TFA) in distilled water, vortexed for about 10 sec, then centrifuged at 4000g for 10 min. The precipitate was discarded and the supernatant filtered through a 0.20 I~ Acro LC 13 filter. A 20 ixL aliquot was injected onto a Protesil 300 Octyl Reverse Phase Column (Whatman, Inc., Clifton, N J, USA) to quantify peak 1 (holo-TTR-RBP).To quantify peak 2b (vitamin A derived from holo-free RBP and from an unidentified high molecular weight lipoprotein associated peak), 120 I~L of the diluted sample preparation was reinjected. Elution of both peaks was with a gradient of buffer A (0.1% TFA in distilled water) and buffer B (0.1% TFA in isopropanol). The gradient was started with 10% buffer B for ! min, then increased linearly to 70% buffer B over 19 min. The gradient stabilized at 70% buffer B for 1 min, then decreased linearly to 10% buffer B for 9 min. The gradient was maintained at 10% buffer B over the last 10 rain of the run (total run time 40 min). The flow rate was a constant 1.0 mL/min, with pH constant at 2.2. The identity of peak 1 was confirmed by repeated chromatography on the TSK molecular exclusion column, by extraction of retinol from the HPLC peaks, and by comparison with human RBP. An aliquot from a human plasma pool of known RBP concentration was run after approximately every tenth rat sample, to correct for possible changes in the column.
Comparison o f TSK and reverse phase H P L C The areas of peak 1 (holo-TTR-RBP) and peak 2a (holo-free RBP) from the TSK 2000 molecular exclusion column were correlated to the areas of peak 1 and peak 2b derived from the reverse phase Protesil Octyl 300 column. Linear correlation coefficients were derived from SAS (Statistical Analysis System, Cary, NC, USA), using the best least squares fit lines.
Quantitation o f holo-RBP Holo-free and holo-TTR-RBP concentrations were estimated by three previously described ~7 independent methods. These methods were: (1) extraction and quantitation of retinol from the HPLC peaks; (2) comparison to purified human vitamin Asaturated free RBP; and (3) comparison of UV detection at 330 nm with fluorescent detection at 330 nm excitation, 418 emission. The estimates from each method were averaged to get the reported RBP concentration.
Estimation o f apo R B P concentrations A small amount of a concentrated solution of retinol in 95% ethanol was added to samples of rat plasma so that the retinol was present in a 4 to 10 times excess molar concentration relative to RBP. These vitamin A-saturated preparations were incubated overnight at 4° C, then run on the TSK HPLC column using standard conditions. Apo RBP concentrations were estimated by subtracting the peak areas of rat plasma incubated with 95% ethanol alone from the peak areas of rat plasma incubated with retinol. Peak areas for the unknown samples were compared to HPLC peak areas of known standards. First order linear regression coefficients were derived with Sigmaplot (Jandel Scientific, Sausalito, CA, USA) and with SAS and were based on the best least squares fit of the data.
Results T h e rats u s e d in t h e first s t u d y h a d v i t a m i n A l i v e r s t o r e s r a n g i n g f r o m m a r g i n a l l y d e f i c i e n t to a d e q u a t e . T h e r a t s in t h e s e c o n d s t u d y w e r e a d e q u a t e o r sub-
toxic. Most rats had adequate liver vitamin A stores at t h e e n d o f t h e e x p e r i m e n t s . A t no t i m e d i d a n y o f t h e s e r a t s s h o w s y m p t o m s o f g r o s s v i t a m i n A defic i e n c y o r t o x i c i t y . A l l r a t s c o n t i n u e d to gain w e i g h t during the study, though the rate of growth was s l o w e d d u r i n g t h e l a s t w e e k t h e r a t s w e r e f e d the uns u p p l e m e n t e d v i t a m i n A - d e f i c i e n t diet. C o a t s , e y e s , and other physical symptoms remained normal. Rat w e i g h t s , l i v e r w e i g h t s , a n d h e m a t o c r i t s w e r e n o t sign i f i c a n t l y d i f f e r e n t in a n y g r o u p o f rats at t h e b e g i n n i n g or end of the times they were on different vitamin A supplements. Liver and plasma vitamin A concentrat i o n s , h o w e v e r , d i d d i f f e r a m o n g g r o u p s . T h e rats g i v e n t h e l o w e s t v i t a m i n A s u p p l e m e n t h a d the l o w e s t l i v e r v i t a m i n A c o n c e n t r a t i o n s (Table 2). R a t h o l o - R B P h a s v e r y s i m i l a r p r o p e r t i e s to h u m a n h o l o - R B P o n m o l e c u l a r e x c l u s i o n H P L C 17 (Figures la and Ib). T h r e e p e a k s a r e p r e s e n t in m o l e c u l a r exc l u s i o n c h r o m a t o g r a m s o f R B P . T h e 17-minute p e a k ( p e a k 1) h a s b e e n i d e n t i f i e d as h o l o - T T R - R B P . 17 T h e 2 4 - m i n u t e p e a k ( p e a k 2a) h a s b e e n identified as h o i o free RBP.17 T h e t h i r d p e a k is i d e n t i f i a b l e in all o f o u r c h r o m a t o g r a m s f r o m rat p l a s m a b u t o n l y s o m e o f o u r chromatograms from human plasma, This peak (peak 0) m i g r a t e s at 12 m i n u t e s a n d a p p e a r s to be a high m o l e c u l a r w e i g h t c o m p l e x that c o n t a i n s v i t a m i n A a n d lipoprotein. T h e s i m i l a r i t y o f rat a n d h u m a n h o l o - R B P on rev e r s e p h a s e c h r o m a t o g r a p h y is e q u a l l y striking, as s h o w n b y c o m p a r i s o n s o f Figures 2a and 2b. T w o p e a k s a r e p r e s e n t in the r e v e r s e d p h a s e c h r o m a t o g r a m o f e i t h e r rat o r h u m a n s e r u m , a n d b o t h a r e in s i m i l a r l o c a t i o n s . T h e m a j o r p e a k , ( p e a k 1), r e a c t s with antib o d i e s to R B P a n d to T T R in h u m a n s e r u m a n d has b e e n i d e n t i f i e d as h o l o - T T R - R B P b e c a u s e o f its f l u o r e s c e n t p r o p e r t i e s , its p h y s i o c h e m i c a l p r o p e r t i e s , a n d its v i t a m i n A c o n t e n t (Table 3). T h e small p e a k ( p e a k 2b) h a s b e e n i d e n t i f i e d as v i t a m i n A d e r i v e d f r o m h o l o - f r e e R B P a n d f r o m the high m o l e c u l a r w e i g h t l i p o p r o t e i n - a s s o c i a t e d c o m p l e x , b a s e d on its f l u o r e s c e n t a n d p h y s i o c h e m i c a l p r o p e r t i e s , its v i t a m i n A c o n t e n t , a n d its l a c k o f r e a c t i v i t y to e i t h e r R B P o r t r a n s t h y r e t i n a n t i b o d i e s in h u m a n s e r u m . T h e r e usually is o n l y a s m a l l a m o u n t o f l i p o p r o t e i n - a s s o c i a t e d v i t a m i n A in h u m a n p l a s m a , so p e a k 2b c o r r e s p o n d s c l o s e l y to p e a k 2a of Figure la and lb ( f r e e - h o l o R B P o n t h e m o l e c u l a r e x c l u s i o n c o l u m n ) in h u m a n s . P e a k 2b in rats h a s high c o r r e l a t i o n s w i t h the high m o l e c u l a r weight lipoprotein associated vitamin A complex ( p e a k 0, Figure la) a n d a l e s s e r c o r r e l a t i o n to f r e e h o l o - R B P ( p e a k 2a, Figure la), b e c a u s e it c o n s i s t s o f vitamin A derived from both complexes. H o l o - T T R - R B P d o e s n o t d i s s o c i a t e in e i t h e r the m o l e c u l a r e x c l u s i o n 17 o r t h e r e v e r s e p h a s e m e t h o d w h e n t h e h o l o - T T R - R B P p e a k is c o l l e c t e d a n d rechromatographed three times. Molecular exclusion HPLC and reverse phase H P L C p e a k a r e a s a r e h i g h l y c o r r e l a t e d to e a c h o t h e r (Figure 3). T h e p e a k a r e a s o f p e a k 1 (identified as h o l o - T T R - R B P ) s h o w e d t h e h i g h e s t c o r r e l a t i o n s bet w e e n t h e s e t w o m e t h o d s (r = 0.96). T h e p e a k a r e a s o f p e a k 2a ( f r e e - h o l o R B P , m o l e c u l a r e x c l u s i o n ) a n d
J. Nutr. Biochem., 1992, vol. 3, January
33
Research Communications Table 2
Physiochemical properties of rats fed different amounts of vitamin A
Mg RP Kg diet
n
0.5 1.0 3.0 10.0 Study 2
5 6 6 6 6
Rat liver weight (final) g
Rat weight (final) 433 422 418 448 514
-+ 58 _+ 63 + 22 -+ 30 _+ 17
14.2 16.9 16.3 16.9 16.7
Liver vitamin A I~g RE g liver
-+ 2.7 -+ 3.8 _+ 18 -+ 1,3 _+ 0.6
26.8 32.8 29.4 67.5 352
+ 16.3 -+ 7.4 _+ 7.4 -+ 10.7 ~ _+ 149 a
RP, retinyl palmitate; RE, retinol and retinyl esters. aSignificant difference, P < 0.05.
2b (free-holo RBP and lipoprotein associated vitamin A, reverse phase) are not as highly correlated in these two methods (r = 0.64). There is a higher correlation between peaks 0 (lipoprotein associated vitamin A, molecular exclusion) and 2b (reverse phase), but only in the rat. This is because reverse phase chromatography disassociates retinol from both holo-free RBP and from the unidentified high molecular weight complex of lipoprotein associated with vitamin A. Apo-free and a p o - T T R - R B P are not present in significant concentrations in these rats. The difference in fluorescence between plasma incubated with ethanol or with saturating amounts of vitamin A ranged from - 6 % to + 14%, and were not dependent on the amount of vitamin A fed to the rats. The effects of liver vitamin A concentration on holo-free and h o l o - T T R - R B P in the first experiment, as measured by molecular exclusion chromatography, is shown in Figure 4. H o l o - T T R - R B P (peak 1) was
positively correlated to liver vitamin A stores (r = 0.79) in marginal or adequate rats. The high molecular weight complex containing lipoprotein and vitamin A is also positively correlated to liver vitamin A (r = 0.79). Holo-free RBP, however, has no significant correlation to liver vitamin A (r = 0.14). These correlations are similar to those derived independently from reverse phase H P L C (Figure 5). H o l o - T T R - R B P has an almost identical correlation with liver vitamin A (r -- 0.81). The correlation of reverse phase peak 2b (hoio-free RBP and the high molecular weight vitamin
c:
8 _= LI-
1
o¢ 0
5'
' 10
' 15
2= U_
0
i
i
i
i
I
I
6
12
18
24
30
36
' 2'o 25 Time (rain)
' 30
3'5
Figure 2a Typical reverse phase chromatogram of rat holo-RBP. Chromatogram is of 20 ILL 1:3 diluted rat plasma, HPLC method is as described in the text.
Time (min)
Figure l e
Typical molecular exclusion (TSK) chromatogram of rat holo-RBP. Chromatogram is of 25 ~L 1:3 diluted rat plasma. HPLC method is as described in text.
c ffl ®
2b IJ _
LI.
0
6
12
18
2'4
3'0
0
Time (rain) Figure l b
Typical molecular exclusion (TSK) chromatogram of human holo-RBP. Chromatogram is of 25 ~L of 1:3 diluted human plasma. HPLC method is as described in text.
34
J. Nutr. Biochem., 1992, vol. 3, January
I
5
I
10
I
15
2'o
30
35
Time (rain)
Figure 2b
Typical reverse phase chromatogram of human holoRBP. Chromatogram is of 100 ILL 1:3 diluted human plasma. HPLC method is as described in the text.
Effect of vitamin A on RBP in rats. Burri et al.
Table 3 Properties of HPLC peaks from normal rat plasma (peak assignment)
Retention time, min Reverse phase Molecular exclusion Molecular mass, Da % of total peak area Reverse phase Molecular exclusion % of total extractable vitamin A in peak Reverse phase Molecular exclusion Excitation max, nm Molecular exclusion Emission max, nm Molecular exclusion
Lipid-protein
holo-TTR-RBP
halo-free RBP
-12.5 Exclusion
17.1 17.8 68,000
24.8 25.5 20,500
-2-8
89-95 23-28
5-11 69-75
-3-7
84-95 82-91
2-11 3-14
365
340
335
465
450
450
A complex) to liver vitamin A is insignificant (r = 0.21), probably reflecting the major contribution of holo-free RBP to this peak. The correlations of holoTTR-RBP to liver vitamin A are higher than the correlation of plasma vitamin A to liver vitamin A in these rats (Figure 6, r = 0.58). The correlations for holoTTR-RBP and retinol with liver vitamin A were not good in adequate-to-subtoxic rats. The correlations of liver retinyl ester concentrations with plasma retinol, holo-TTR-RBP and holo-free RBP were 0.26, 0.34, and 0.27, respectively. Preliminary data on vitamin A-deficient humans (plasma retinol 0.2 to 0.4 ~mol/L) showed that both holo-TTR-RBP and free-holo RBP were decreased significantly (P -< 0.01). Serum retinyl esters increased significantly (P -< 0.001) in humans with vitamin A toxicity. Serum retinol, holo-TTR-RBP and free-holo RBP did not increase significantly from the normal human range. Typical molecular exclusion chromatograms for human plasma RBP in deficiency and toxicity are shown in Figures 7a and 7b. Discussion
ing molecular exclusion chromatography on a TSK 2000 column, the other reverse phase chromatography on a Protesil Octyl 300 column. These methods give very similar results for the effect of liver and plasma vitamin A concentrations on holo-TTR-RBP. Molecular exclusion chromatography may be the superior method of the two because it does not disassociate vitamin A from halo-free RBP and from the high molecular weight vitamin A-containing complex associated with lipoprotein, and it requires less sample preparation. Neither method is species specific, and both can assay vitamin A-containing halo-free and holoTTR-RBP. The HPLC methods described are not strictly comparable to immunologic methods, because the HPLC methods measure holo-RBP only, while immunologic methods measure apo-RPB and any other protein with cross-reactivity to RBP antibodies. Both the HPLC and immunologic methods should usually give almost the same total RBP concentrations (within experimental error) because apo-RBP is present in very small amounts in plasma and serum. Our results showed differences of - 6 % to 14% between vitamin A-saturated and non-saturated plasmas in these rats,
We developed two independent HPLC methods for measuring holo-free and holo-TTR-RBP, one involv2000-
5
0
=
Transthyretin-bound
•
=
Free RBP (Peak
RBP
(Peak 1)
2a) O
0 P e a k 1 RP vs Peak 1 TSK P e a k 2 b RP vs Peak 20 TSK
• 1500
& P e a k 1 RP vii P e a k 0 TSK "6
3
1000
o.. 1-
r - 0.635 •
~
O~
o
5oo
J
j •
0
~
O 009..1 ~
oO~.o
O
r - 0.923 A . " .. " ~ . . .A ~ . -
o o
. .........
"
r = 0.794
=E
3
~
2 0
1
~'~''&
.....
~A.~'~,
0 0
600
1200
Reverse
1800
Phase
Peak
2400
3000
Areas
Figure 3 Correlation of peak areas determined by molecular exclusion HPLC with peak areas determined by reverse phase. Methods as described in text.
o. • • ~
~ I
0
• .oqk.~.e.
20
om ....
~
....
•
.......
~
40 60 Uver Retinyl Esters (ug/g)
••
BO
Figure 4 Correlation of liver vitamin A concentration with halofree and holo-TrR-RBP concentrations in rat plasma, as measured by molecular exclusion HPLC. Methods are described in the text. Each point represents the mean of duplicate assays. © = peak 1 (holo-TTR-RBP). • = peak 2a (halo-free RBP). J. Nutr. B i o c h e m . ,
1 9 9 2 , v o l . 3, J a n u a r y
35
Research Communications 6. O - Tr~anmthymtin-bound RBP (Peak 1) • - Retlnol f r o m f r e e R B P a n d Lipoprotein--=ssociated Vitamin A (Peak 2b)
5 i
O
4 o ~
0
tz
2
o
1
~
O
O m
0 ~
i
V
~
0
r = 0.213
. = mmlL •
•
mmmrR~
mm
20
~m
m
m
~
40
~
60
--
80
U v e r Retinyl Esters ( u g / g )
Figure 5 Correlation of liver vitamin A concentration with h a l o free- and holo-TTR-RBP concentrations in rat plasma, as measured by reverse phase HPLC. Methods are described in the text Each point represents the mean of duplicate assays. © = peak 1 (holo-TTR-RBP). • = peak 2b (vitamin A from halo-free RBP and prepeak)
were always run with at least one human plasma control, an accurate comparison of peak areas of typical chromatograms of RBP from rats and humans was possible. This comparison shows that the unidentified high molecular weight complex associated with lipoprotein and vitamin A (peak 0, molecular exclusion) and holo-TTR-RBP (peak l) concentrations were consistently higher in rats than in humans with similar plasma vitamin A concentrations. Holo-free-RBP concentrations were lower in rats, particularly at higher plasma vitamin A concentrations. There are two probable explanations for these differences: (1) the formation or degradation of the holo-TTR-RBP complex is regulated by different mechanisms in rats
Deficiency 2500
,
2500
2000
2000
1500
1500
5. o
~-4-
o
E ~3.
j
r=0
~ 2-
o
8
o o o
o
qb o
o o
,
1000 PIs~
~
I
,°°° i
500
500
-500
-500
o
o
o o
o 0
.....
0
, .....
20
, .....
40
, .....
60
, .....
80
100
U v e r Retinyl P a l m i t a t e (ug / g)
Figure 6 Correlation of liver vitamin A concentration with plasma vitamin A concentration. Liver retinot esters and plasma retinal were determined by isocratic HPLC as described in text All assays were done in d u p l i c a t e
with no consistent effect of vitamin A concentration in the diet. This study shows holo-TTR-RBP concentrations were better than plasma retinal concentrations as indicators of liver vitamin A stores in marginally deficient-to-normal rats. The correlations between holo-TTR-RBP and liver vitamin A concentrations were significantly better (P = 0.01) than the correlation of plasma vitamin A concentrations to liver vitamin A concentrations. This suggests that holo-TTRRBP concentrations may be influenced by vitamin A liver stores to a greater extent than plasma vitamin A is in marginally deficient rats, and may be better indicators of vitamin A status in normal rats. Preliminary human data suggests that serum retinal, halo-free RBP and holo-TTR-RBP all decrease significantly in vitamin A deficiency. Data from humans and rats suggests that these relationships break down during vitamin A subtoxicity or toxicity. In toxicity, retinyl esters were increased significantly in humans, but serum retinal, holo-TTRRBP and halo-free RBP concentrations were not. Our rat and human chromatograms always showed characteristic differences. Since rat plasma samples
36
J Nutr Biochem., 1992, vol. 3, January
-1000
10
0
20
30
40-1000
Molecular exclusion chromatogram of human haloRBP in vitamin A deficiency Chromatogram is of 25 IxL of 1:3 diluted plasma• HPLC method is as described in text Figure 7a
Toxicity 2500
2500
i
2000
2000
Plwk t 1500
1500
•
pe.L.
1000
•
|
500 ~
:2 -500
t
-1000
0
1'0
0
-500
70
40-- 1000
(.~.ntJm)
Molecular exclusion chromatogram of human haloRBP in vitamin A toxicity. Chromatogram is of 25 ixL of 1:3 diluted plasma. HPLC method is as described in text. Fibure 7b
Effect of vitamin A on RBP in rats: Burri et al.
and humans; (2) the formation of the holo-TTR-RBP complex is influenced by nutritional factors other than vitamin A status. Little is known about the factors influencing the formation of the holo-TTR-RBP complex in vivo, but there is some support for both possibilities. Rat and human RBP have different amino acid compositions, ~8 and different immunologic reactivities. ~9Because RBP seems to be different in these species, it would not be surprising if the binding of RBP to TTR was different. It is also true that the typical diets of rats and humans are different. Human data suggested that free RBP is affected to a much greater extent by changes in liver function (shown in subjects with fluctuating creatinine) or by daily changes in nutrient intake (as indicated by greater day-to-day variations in the free RBP peak. 2° It is possible that RBP levels in this study may have been influenced by differences in dietary factors such as zinc, 2~ protein or calories 1°-12or carbohydrate, z: The rats used in these studies have liver vitamin A stores which were marginally deficient to subtoxic. This narrow range of vitamin A nutriture was used in an attempt to mimic the normal range of vitamin A nutriture in the human populations of the United States, where gross vitamin A deficiency or toxicity is uncommon, but marginal status occurs fairly often. 23 It appears that holo-TTR-RBP may be a good indicator of vitamin A status in this range of vitamin A nutriture. Unfortunately, it is unclear whether the rat is a good model for humans, since it is possible that the binding of TTR to RBP is regulated by different mechanisms in these species. The mechanisms controlling the formation of the holo-TTR-RBP complex require more study than they have been given. The molecular exclusion HPLC method described should provide a useful tool for studying this complex and the mechanisms controlling its formation and degradation in animals and man.
Acknowledgments We would like to thank Peter Taylor, Pearl Sun, and Denise Gretz for their technical assistance. We would also like to thank Drs. Robert Russell, Daniel Bankson, Harold Zwick, and Mary Kretsch for kindly donating blood samples. References 1 2 3 4 5
Goodman, D.S. (1984). Plasma retinol-binding protein. In The Retinoids. (A.B. Sporn and D.S. Goodman, eds.) vol. 2, pp. 41-88, Academic Press, Orlando, FL Goodman, D.S. (1980). Plasma retinol-binding protein. Ann. N . Y . Acad. Sci. 348, 378-390 Muto, Y., Smith, J.E., Milch, P.O., and Goodman, D.S. (1972). Regulation of retinol-binding protein metabolism by vitamin A status in the rat. J. Biol. Chem. 247, 2542-2550 Mallia, A.K., Smith, J.E., and Goodman, D.S. (1975). Metabolism of retinol-binding protein and vitamin A during hypervitaminosis in the rat. J. Lipid Res. 16, 180-188 Harrison, E.H., Smith, J.E., and Goodman, D.S. (1980). Effects of vitamin A deficiency on the levels and distribution of
6
7 8
9 10
11
12
13 14 15 16 17
18 19 20
21
22
23
retinol-binding protein and marker enzymes in homogenates and Golgi-Rich fraction of rat liver. Biochimica et Biophysica Acta 628, 489-497 Audisio, M., Dante, D., Fidanza, F., Villani, C., and Rulli, G. (1985). Valori plasmatici della vitamina A e dei suoi vettori proteici (rbp e PA), del corotene, del colesterolo total e del colesterolo-HDL in soggetti affetti da carcinoma mammario. Boll. Soc. It. Biol. Sper. 61,467-474 Smith, F.R. and Goodman, D.S. (1971). The effects of diseases of the liver, thyroid, and kidneys on the transport of vitamin A in human plasma. J. Clin. Invest. 50, 2426-2436 Smith, F.R., Suskind, R. Thanangkul, O., Leitzmann, C., Goodman, D.S., and Olson, R.E. (1975). Plasma vitamin A, retinol-binding protein and prealbumin concentrations in protein-calorie malnutrition. A m J. Clin. Nutr. 28, 732-738 Smith, F. R., Underwood, B., Denning C., Varma, A., and Goodman, D.S. (1972). Depressed plasma retinol binding protein levels in cystic fibrosis. J. Lab. Clin. Med. 80, 423-433 Smith, F.R., Goodman, D.S., Zaklama, M.S.. Gabr, M., Maraghy, S.E., and Batwardhan, W. (1973). Serum vitamin A, retinol-binding protein, and prealbumin concentration in protein calorie malnutrition. I. Functional defect in hepatic retinol release. Am. J. Clin. Nutr. 26, 973-981 Smith, F.R., Goodman, D.S., Arroyave, G., and Viteri, F. (1973). Serum vitamin A, retinol-binding protein, and prealbumin concentrations in protein-calorie malnutrition. 11. Treatment including supplemental vitamin A. Am. J. Clin. Nutr. 26, 982-987 lngenbleek, Y., Van den Schrieck, H.G., De Nayer, P., and De Visscher, M. (1975). Albumin, transferrin and the thyroxine-binding prealbumin/retinol-binding protein (TBPARBP) complex in assessment of malnutrition. Clin. Chim. Acta 63, 61-67 Fex, G. and Felding, P. (1984). Factors affecting the concentration of free holo retinol-binding protein in human plasma. Eur. J. Clin. Invest. 14, 146-149 Catignani, G.L. and Bieri, J.G. (1983). Simultaneous determination of retinol and c~-tocopherol in serum or plasma by liquid chromatography. Clin. Chem. 29, 708-712 Chow, F.I. and Omaye, S.K. (1983). Use of antioxidants in the analysis of vitamins A and E in mammalian plasma by high performance liquid chromatography. Lipids. 18, 837-841 National Research Council (1978). Nutrient requirements of laboratory animals, vol. 10, 3rd ed., pp. 7-37. National Research Council, Washington, D.C. Burri, B.J. and Kutnink. M.A. (1989). Liquid-chromatographic assay for free and transthyretin-bound retinolbinding protein in serum from normal humans. Clin. Chem. 35, 582-586 Rask, L., Anundi, H., and Peterson, P.A. (1979). The primary structure of the human retinol-binding protein. FEBS Letters 104, 55-58 Muto, Y., Smith, F.. and Goodman, D.S. (1973). Comparative studies of retinol transport in plasma. J. Lipid Res. 14, 525-532 Burri, B.J., Bankson, D.C., and Neidlinger, T.R. (1990). Use of free and transthyretin-bound retinol-binding protein in serum as tests of vitamin A status in humans: Effect of high creatinine concentrations in serum. Clin. Chem. 36, 674-676 Shingwekar, A.G., Mohanram, M., and Reddy, V. (1979). Effects of zinc supplementation on plasma levels of vitamin A and retinol-binding protein in malnourished children. Clin. Chem. Acta 93, 97-100 Kelleher, P.C., Phinney, S.D., Sims, E.A.H., Bogardus, C., Horton, E.S., Bistrian, B.R., Amatruda, J.M., and Lockwood, D.H. (1983). Effects of carbohydrate-containing and carbohydrate-restricted hypocaloric and eucaloric diets on serum concentrations of retinol-binding protein, thyroxinebinding prealbumin and transferrin. Metabolism 32, 95-101 Pilch, S.M. (1985). Assessment of the vitamin A nutritional Status of the U.S. Population based on data collected in the health and nutrition examination surveys. LSRO, FASEB Bethesda, MD
J. Nutr. Biochem., 1992, vol. 3, January
37