- Email: [email protected]
Plasma holotranscobalamin compared with plasma cobalamins for assessment of vitamin B12 absorption; optimisation of a non-radioactive vitamin B12 absorption test (CobaSorb)
Clinica Chimica Acta 376 (2007) 150 – 154 www.elsevier.com/locate/clinchim
Plasma holotranscobalamin compared with plasma cobalamins for assessment of vitamin B12 absorption; optimisation of a non-radioactive vitamin B12 absorption test (CobaSorb) Anne-Mette Hvas a,1 , Anne L. Morkbak b,⁎,1 , Ebba Nexo b a
b
Department of Clinical Biochemistry, Skejby Hospital, Aarhus University Hospital, 8000 Aarhus, Denmark Department of Clinical Biochemistry, Aarhus Hospital, Aarhus University Hospital, Norrebrogade 44, DK- 8000 Aarhus, Denmark Received 12 June 2006; received in revised form 7 August 2006; accepted 7 August 2006 Available online 11 August 2006
Abstract Background: A recently developed non-radioactive vitamin B12 absorption test (CobaSorb) was further explored to identify the best marker for reflection of vitamin B12 absorption and to determine the duration of the test. Methods: Seventy-eight healthy individuals (age 21–81 years) were given three oral doses of 9 μg vitamin B12 per day for 5 successive days. Non-fasting blood samples were collected on days 1 to 5 before administration of vitamin B12 and on day 8. Cobalamins and holotranscobalamin were measured. Results: Performance of the vitamin B12 absorption test was evaluated in individuals with holotranscobalamin or cobalamins below the 75% percentiles. We used a change greater than 2 × CVday-to-day in holotranscobalamin (22%) and cobalamins (12%) to indicate a change caused by absorption of vitamin B12. Among individuals with a baseline holotranscobalamin below the 75% percentile (b 75 pmol/L, n = 57), 98% had an increase in holotranscobalamin N 22% from day 1 to day 3. In contrast, only 72% of the individuals with baseline cobalamins below the 75% percentile (b 335 pmol/L, n = 57) had an increase in cobalamins N 12%. Conclusions: In healthy individuals with baseline holotranscobalamin b 75 pmol/L, vitamin B12 absorption is well reflected by an increase in holotranscobalamin after 2 days administration of oral vitamin B12. © 2006 Elsevier B.V. All rights reserved. Keywords: Absorption test; Cobalamin; Holotranscobalamin; Vitamin B12
1. Introduction Vitamin B12 is an essential micronutrient supplied by meat or dairy products. Vitamin B12 is a co-enzyme in one-carbon metabolism and essential for DNA synthesis. The clinical consequences of vitamin B12 deficiency include megaloblastic anemia and neurological disorders. In the stomach, food-bound vitamin B12 is released from dietary protein and transferred to the small intestine in complex with haptocorrin (HC). In the alkaline environment of the small intestine, HC is degraded by pancreatic enzymes. The free vitamin B12 forms a complex with intrinsic Abbreviations: HoloTC, holotranscobalamin; IF, intrinsic factor; TC, transcobalamin; HC, haptocorrin; tHcy, total homocysteine; MMA, methylmalonic acid. ⁎ Corresponding author. Tel.: +45 8949 3065; fax: +45 8949 3060. E-mail address: [email protected] (A.L. Morkbak). 1 Shared first authorship. 0009-8981/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2006.08.009
factor (IF), a protein produced in the gastric parietal cells, and the IF–vitamin B12 complex is absorbed by receptor-mediated uptake in the ileum [1–3]. If no IF is available which is seen in some patients after gastrectomy or in patients with pernicious anemia, only around 1% of the ingested vitamin B12 is taken up by passive diffusion [4]. When IF is available, approximately 75% of a single physiological dose less than 0.5 μg is absorbed and about 50% of a dose around 1.0 μg [4]. For a dose around 5 μg, the average uptake is 25%, with no further increase of the total vitamin B12 absorbed if the oral concentration is increased [4]. In the circulation, vitamin B12 is bound to two proteins, transcobalamin (TC) and HC. The TC-vitamin B12 complex (holoTC) is transported from the circulation into the cells where vitamin B12 is released after degradation of TC [5]. The physiological role of HC is still unsettled. After diagnosing vitamin B12 deficiency, the cause of deficiency has to be clarified. For many years, the Schillings test was used for
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154
this purpose [6]. In the test, labelled vitamin B12 without (Schillings test I) or with IF (Schillings test II) is administered orally. Afterwards, a large dose of vitamin B12 is injected in order to saturate endogenous vitamin B12 binders and thereby allowing the absorbed vitamin to be excreted in the urine. The amount of labelled vitamin B12 excreted in a 24-h urine sample mirrors the capacity to absorb vitamin B12. However, the cumbersome Schillings test has almost disappeared from the market for several reasons, i.e., difficulties in obtaining labelled vitamin B12 as well as native human IF for Schillings test II no longer being available. Substitution for human IF such as hog IF is difficult to obtain and is used only sparsely due to safety concerns. Thus, there is a need for alternative tests for determination of the vitamin B12 absorption. Recently, a new vitamin B12 absorption test, CobaSorb, based on oral ingestion of three 9 μg doses of non-radioactive vitamin B12 at 6-h intervals was introduced and evaluated preliminarily in healthy subjects [7] and in patients with Imerslund-Grasbeck syndrome or inherited lack of intrinsic factor [8]. After ingestion of vitamin B12, a significant increase in holoTC was seen in the healthy individuals [7,8], whereas holoTC remained unchanged in the patients [8]. In the present study, we further explored the design of CobaSorb.
151
on Biomedical Research Ethics of Aarhus County (Project Number 2005-0067).
2.2. Study design Vitamin B12 absorption was evaluated by analysis of cobalamins and holoTC in samples obtained before, during, and after intake of oral vitamin B12. The individuals took three oral doses of 9 μg cyanocobalamin (Natur-Drogeriet, Horning, Denmark) (vitamin B12) per day for 5 successive days (day 1 to 5). Vitamin B12 was administered at 6-h intervals (0800, 1400, and 2000 h), together with orange or apple juice and a snack (fruit, biscuit or bread). Non-fasting blood samples were collected on days 1 to 5 before administration of the first dose of vitamin B12 and in the morning on day 8. 2.3. Sampling and analytical methods Serum was collected for analysis of cobalamins, holoTC, and total TC (days 1 to 5 and day 8), methylmalonic acid (MMA) (day 1 and day 8) and autoantibodies against IF (day 1). EDTA blood was collected for analysis of hemoglobin and erythrocyte mean cell volume (EMV) (day 1). Lithium heparin plasma was drawn for analysis of creatinine (day 1). Heparin fluoride plasma was collected for analysis of total homocysteine (tHcy) (day 1 and day 8). Plasma for analysis of tHcy was separated from the blood cells within 2 h after sample collection and stored at − 80 °C until analysed. Cobalamins were assayed on the Advia Centaur (Bayer A/S, Germany) with a total imprecision b 10% (intra-assay imprecision b 5%). Total TC was measured by an in-house sandwich ELISA with an imprecision of 4% to 6% (intra-assay imprecision ∼ 3%) [9]. HoloTC was measured by the TC ELISA after removal of the apoTC with vitamin B12 coated beads [10]. The total imprecision was ∼ 8% [10], and the intra-
2. Materials and methods 2.1. Study population In total, 78 healthy individuals aged ≥ 18 years participated in the study. Exclusion criteria were vitamin B12 injections given within the past 5 years, use of vitamin pills containing N 1 μg vitamin B12 within the past 3 weeks, infectious disease, pregnancy or lactation, giving birth within the past 9 months, use of hormones/other systemic medication, or known chronic systemic disease. Written informed consent was obtained from all individuals, and the study protocol was approved by the Regional Committee
Table 1 Concentration of cobalamins, holotranscobalamin (holoTC), total TC, TC saturation (days 1 to 5 and day 8), serum methylmalonic acid (MMA), plasma total homocysteine (tHcy) (day 1 and 8) and IF antibodies, erythrocyte mean cell volume (EMV), hemoglobin and creatinine (day 1)
Cobalamins (pmol/L) [200–600a] HoloTC (pmol/L) [40–150b] Total TC (pmol/L) [610–1400c] TC saturation [0.05–0.20b] MMA (μmol/L) [0.08–0.28d] tHcy (μmol/L) [4.5–11.9d] IF antibodies EMV (fL) [85–100] Hemoglobin (mmol/L) Females [7.4–9.6] Males [8.4–10.8] Creatinine (μmol/L) Females [44–115] Males [62–133]
Day 1
Day 2
Day 3
Day 4
Day 5
Day 8
272 (151–528) 54 (19–133) 685 (446–1221) 0.08 (0.04–0.2) 0.1 (0.07–0.4) 7.9 (4.5–22.3) 0 91 (72–101)
298 (160–529) 78 (23–186) 651 (433–1216) 0.12 (0.04–0.3) – – – –
312 (188–648) 89 (33–186) 649 (441–1081) 0.14 (0.05–0.3) – – – –
318 (191–674) 96 (30–213) 639 (444–1073) 0.15 (0.05–0.3) – – – –
343 (201–664) 98 (29–215) 629 (478–1170) 0.16 (0.05–0.3) – – – –
309 (165–548) 62 (21–226) 654 (471–1085) 0.1 (0.04–0.3) 0.1 (0.07–0.3) 8.3 (4.2–34.5) – –
8.4 (6.9–9.5) 9.2 (7.5–10.4)
– –
– –
– –
– –
– –
68 (48–82) 81 (65–110)
– –
– –
– –
– –
– –
Median and (range) are indicated. [95% prediction intervals] are from aNexø [15]; bNexø et al. [10]; cNexø et al. [9]; and dRasmussen K. et al. [16]. The reference interval for tHcy is 4.6–8.1 μmol/L for subjects b30 years, 4.5–7.9 μmol/L for females and 6.3–11.2 μmol/L for males 30–59 years and 5.8–11.9 μmol/L for subjects N60 years.
152
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154
Table 2 Characteristics of the three individuals with serum methylmalonic acid (MMA) N0.28 μmol/L Individual
Age (years) Sex Cobalamins (pmol/L) Delta cobalamins (pmol/La) Total TC (pmol/L) HoloTC (pmol/L) Delta holoTC (pmol/La) TC saturation MMA (μmol/L) Delta MMA (μmol/Lb) tHcy (μmol/L) EMV (fL) Hemoglobin (mmol/L) Creatinine (μmol/L)
1
2
3
60 F 211 31 672 60 17 (28%) 0.09 0.29 −0.02 13.7 98 8.6 72
73 F 176 30 531 19 14 (74%) 0.04 0.30 − 0.04 14.2 91 8.9 57
53 F 172 25 662 31 13 (42%) 0.05 0.43 − 0.33 4.5 96 8.7 65
a
Increase from day 1 to day 3. Decrease in MMA from day 1 to day 8.
on the IMMULITE 2000 (Diagnostic Products Corporation, California) (total imprecision b 6%) [14]. Measurement of tHcy (day 1) failed in one individual. Hemoglobin and EMV were assessed on the Sysmex XE2100 (Sysmex, Japan) with a total imprecision of 2%. Creatinine was measured by the protein-compensated Jaffe method or the enzymatic method on the Roche Cobas Integra 800 analyzer (Roche Diagnostics, Indianapolis) with a total imprecision b 3%. 2.4. Statistical analysis The associations between age, sex and the biochemical parameters were analysed by using linear regression. Changes in means and medians were compared by paired t-test and Mann–Whitney test, respectively. To fulfil the criteria for Gaussian distribution, data were log-transformed when appropriate. P-values b 0.05 were regarded as statistically significant. All statistics were performed in Stata 8.0 (StatCorp LP, Texas).
b
3. Results assay imprecision was ~ 4% [11]. The levels of cobalamins and TC (holo and total) on days 1 to 5 and day 8 for each individual were measured in the same assay. TC saturation was calculated as holoTC/total TC. The presence of autoantibodies against IF was determined by an in-house ELISA employing immobilized recombinant IF as catching reagent and biotinylated recombinant IF as detection reagent [12]. MMA was measured by slightly modified stable-isotopedilution capillary gas chromatography mass-spectrometry (total imprecision b 8%) [13] and tHcy by the immunological method
Among the 78 individuals, 40 were males and 38 were females. The median age was 51 years, range 21–81 years, (median (range) for men was 50 years (22–81 years) and for women 52 years (21–74 years)). The biochemical characteristics are shown in Table 1. All but three individuals had normal vitamin B12 status as judged from a level of MMA within the reference interval (0.08–0.28 μmol/L [16]). Two of these individuals had low cobalamins (reference interval 200– 600 pmol/L [15]) and a low holoTC (reference interval 40– 150 pmol/L [10]), Table 2.
Fig. 1. Baseline holotranscobalamin (holoTC) versus delta holoTC and baseline cobalamins versus delta cobalamins from day 1 to day 2, day 1 to day 3 and day 1 to day 5. Broken lines indicate 2 × coefficient of variation (day-to-day) of the baseline levels of holoTC (22%) and cobalamins (12%). Individuals with a baseline holoTC ≤ 75 pmol/L or cobalamins ≤ 335 pmol/L (•); holoTC N 75 pmol/L or cobalamins N 335 pmol/L (•).
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154 Table 3 Increase in holoTC or cobalamins from day 1 to day 2, 3, 4, 5 and 8 for individuals with a holoTC ≤75 pmol/L (n = 57) or cobalamins ≤335 pmol/L (n = 57) Days
1–2 1–3 1–4 1–5 1–8
Delta holoTC (pmol/L)
Delta cobalamins (pmol/L)
Median (range) [fraction]
Median (range) [fraction]
21 (−10–73) [88%] 32 (1–61) [98%] 36 (7–84) [98%] 39 (10–106) [100%] 5 (−16–50)
30 (−39–85) [53%] 39 (−3–107) [72%] 55 (−20–141) [81%] 65 (−4–146) [86%] 35 (−46–119)
Fraction of individuals with a delta holoTC N22% or delta cobalamins N12% relative to the baseline level.
Using linear regression, we found a significant positive association between age and baseline holoTC (P = 0.03) as well as TC saturation (P = 0.04). No association was found between age and baseline total TC (P = 0.84), cobalamins (P = 0.46), MMA (P = 0.54) or tHcy (P = 0.38), respectively. Sex was not significantly associated with baseline values of holoTC (P = 0.83), cobalamins (P = 0.55), MMA (P = 0.07) or tHcy (P = 0.13). After adjusting for age and sex, a small but significant positive association was observed between baseline levels and increases in holoTC (P = 0.04) and cobalamins (P = 0.04) from days 1 to 5. Interestingly, we found a significant negative association between age and increase in holoTC (P = 0.004) from days 1 to 5. The median (range) increase in holoTC was 42 pmol/L (−2– 109 pmol/L) for individuals ≤ 50 years (n = 39) and 32 pmol/L (− 11–84) for individuals N50 years (n = 39) (P = 0.006). Age was not significantly associated with increase in cobalamins from days 1 to 5 (P = 0.06). No significant association was found between sex and increases in holoTC (P = 0.13) and cobalamins (P = 0.06). Because sex was not associated with an increase in holoTC or cobalamins and the alteration in delta holoTC with age was numerically small, we decided to analyse the total study population together. From a previous study, we know that the day-to-day variation is 11% for holoTC and 6% for cobalamins [7]. We used a change larger than 2 × coefficient of variation (day-to-day) in holoTC (22%) and cobalamins (12%) to indicate a change caused by absorption of the administered vitamin B12. As shown in Fig. 1, a substantial number of individuals with baseline levels above the 75% percentile for holoTC (75 pmol/L) or cobalamins (335 pmol/L) did not show a significant increase in holoTC (6 of 21 individuals (− 16–20%)) or cobalamins (15 of 21 individuals (− 3–12%)). Because a vitamin B12 absorption test is relevant only in individuals with borderline or low levels of holoTC or cobalamins, we analysed in detail the pattern of absorption only in those having baseline levels below the 75% percentiles. Among individuals with a baseline holoTC ≤ 75 pmol/L (b 75% percentile, n = 57), 50 (88%) had an increase in holoTC N 22% on day 2 and 56 (98%) on day 3. Among individuals with baseline cobalamins ≤335 pmol/L (b 75% percentile, n = 57), 30 (53%) had an increase in cobalamins N12% on day 2 and 41 (72%) on day 3, Table 3. The increment in both parameters was largest from day 1 to day
153
2. After day 3, the changes in holoTC and cobalamins were very modest, Table 3. All together, the median increase in holoTC from day 1 to day 3 was 55%, with a 95% prediction interval of 28–146% in individuals with a baseline holoTC ≤75 pmol/L. From day 5 to day 8, the median (range) decrease was − 30 pmol/L (− 85–15 pmol/L) for holoTC and − 35 pmol/L (− 164–39 pmol/L) for cobalamins in individuals with a baseline holoTC ≤ 75 pmol/L. On day 8, however, 2.5 days after the last dose of vitamin B12, the levels of holoTC (P b 0.0001) and cobalamins (P b 0.0001) were still higher than baseline levels. MMA was significantly lower on day 8 compared with day 1 (P = 0.001), whereas this was not the case for tHcy (P = 0.71). The group of individuals with a baseline holoTC N 75 pmol/L showed no significant decrease in MMA (P = 0.61) from day 1 to day 8. 4. Discussion In this study, changes in cobalamins and holoTC were compared regarding suitability as markers of vitamin B12 absorption. Moreover, the design of the vitamin B12 absorption test CobaSorb [7] was further explored by prolonging the duration of vitamin B12 administration. In accord with previously published results [7,8], and as expected theoretically, we found that changes in holoTC reflected vitamin B12 absorption better than changes in cobalamins. The absolute increases in cobalamins and holoTC were comparable, suggesting that the increase in cobalamins is caused by an increase in holoTC and not in holoHC. The increase in the level of cobalamins relative to baseline was low and often within the expected variation of the baseline level (Fig. 1). Therefore, we will only discuss changes in holoTC and not in cobalamins in relation to absorption of vitamin B12. We found that increased age was associated with a smaller increment in holoTC after vitamin B12 ingestion, suggesting that the ability to absorb crystalline vitamin B12 decreases with age. This was also observed in a smaller, previously published
Fig. 2. Practical use of the vitamin B12 absorption test, CobaSorb.
154
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154
study based on CobaSorb [8]. A decline in absorption of vitamin B12 with age can be explained by a decrease in the production of IF or reduced absorptive capacity or both. In contrast to our results, a study based on the conventional Schillings test suggests no decrease in absorption of crystalline B12 with age [17]. In this study, however, only elderly persons were examined and the results were compared with previously published references intervals. Furthermore, the imprecision of the Schillings test is relatively large, and thus, an age-related decline may have been overlooked in this study. Since evaluation of vitamin B12 absorption is of interest only in patients suspected of having vitamin B12 deficiency, often based on borderline or abnormal markers for vitamin B12 status, we found it most relevant to scrutinize results from the individuals with a baseline holoTC below the 75% percentile (b75 pmol/L). These individuals showed a marked increase in holoTC one day after the start of oral vitamin B12 and a further increase the following days. Though a new steady-state was not established after 4 days of vitamin B12 administration, the increase in holoTC levelled off after 2 days. Therefore, we recommend that future testing employing the CobaSorb design include a baseline blood sample for measurement of holoTC followed by 2 days administration of three times 9 μg vitamin B12 and subsequent collection of a second sample for analysis of holoTC. Based on our data, we believe that an increase in holoTC of at least 22% supports the patient being capable of absorbing vitamin B12. Because the analytical detection limit for holoTC is 6.4 pmol/L, we additionally require an increase in holoTC ≥ 10 pmol/L. The practical use of CobaSorb is outlined in Fig. 2. A limitation of this study is that it reports results only on healthy individuals and does not include data on patients with mal-absorption of vitamin B12. It has already been documented that individuals unable to absorb vitamin B12 does not increase in holoTC one day after administration of the vitamin [8,18], but even though it is unlikely, it has to be proven that holoTC does not increase significantly in mal-absorbers after 2 days of administration of three times 9 μg vitamin B12. In conclusion, we report that in healthy individuals with baseline holoTC below the 75% percentile (b 75 pmol/L) vitamin B12 absorption is reflected by an increase in holoTC– but not cobalamins–after 2 days' administration of oral vitamin B12, and we suggest that future use of CobaSorb is done based on this design. Acknowledgement We thank Jette Fisker Petersen and Anna-Lisa Christensen at Department of Clinical Biochemistry, Aarhus Sygehus, and Lisbeth Norengaard Sørensen and Kirsten Christiansen at Department of Clinical Biochemistry, Center of Hemophilia
and Thrombosis, Skejby Hospital, Aarhus University Hospital, Denmark, for outstanding technical assistance. References [1] Moestrup SK, Verroust PJ. Megalin- and cubilin-mediated endocytosis of protein-bound vitamins, lipids, and hormones in polarized epithelia. Annu Rev Nutr 2001;21:407–28. [2] Kristiansen M, Kozyraki R, Jacobsen C, Nexo E, Verroust PJ, Moestrup SK. Molecular dissection of the intrinsic factor-vitamin B12 receptor, cubilin, discloses regions important for membrane association and ligand binding. J Biol Chem 1999;274:20540–4. [3] He Q, Madsen M, Kilkenney A, et al. Amnionless function is required for cubilin brush-border expression and intrinsic factor-cobalamin (vitamin B12) absorption in vivo. Blood 2005;106:1447–53. [4] Chanarin I. The megaloblastic anaemias. Third ed. London: Blackwell Scientific Publications; 1990. [5] Seetharam B, Yammani RR. Cobalamin transport proteins and their cellsurface receptors. Expert Rev Mol Med 2003;5:1–18. [6] Schilling RF. Intrinsic factor studies II. The effect of gastric juice on the urinary excretion of radioactivity after the oral administration of radioactive vitamin B12. J Lab Clin Med 1953;42:860–5. [7] Bor MV, Nexo E, Hvas AM. Holo-transcobalamin concentration and transcobalamin saturation reflect recent vitamin B12 absorption better than does serum vitamin B12. Clin Chem 2004;50:1043–9. [8] Bor MV, Cetin M, Aytac S, Altay C, Nexo E. Nonradioactive vitamin B12 absorption test evaluated in controls and in patients with inherited malabsorption of vitamin B12. Clin Chem 2005;51:2151–5. [9] Nexo E, Christensen AL, Petersen TE, Fedosov SN. Measurement of transcobalamin by ELISA. Clin Chem 2000;46:1643–9. [10] Nexo E, Christensen AL, Hvas AM, Petersen TE, Fedosov SN. Quantitation of holo-transcobalamin, a marker of vitamin B12 deficiency. Clin Chem 2002;48:561–2. [11] Morkbak AL, Heimdal RM, Emmens K, et al. Evaluation of the technical performance of novel holotranscobalamin (holoTC) assays in a multicenter European demonstration project. Clin Chem Lab Med 2005;43:1058–64. [12] Nexo E, Nooroya B, Hvas AM, Christensen AL, Waters H. Autoantibodies against intrinsic factor (IF) measured with an ELISA using recombinant human IF as both catching and detecting reagent. Clin Chem Lab Med 2005;43:351–6. [13] Rasmussen K. Solid-phase sample extraction for rapid determination of methylmalonic acid in serum and urine by a stable-isotope-dilution method. Clin Chem 1989;35:260–4. [14] Moller J, Ahola L, Abrahamsson L. Evaluation of the DPC IMMULITE 2000 assay for total homocysteine in plasma. Scand J Clin Lab Invest 2002;62:369–73. [15] Nexø E. Variation with age of reference values for P-cobalamins. Scand J Haematol 1983;30:430–2. [16] Rasmussen K, Moller J, Lyngbak M, Pedersen AM, Dybkjaer L. Age- and gender-specific reference intervals for total homocysteine and methylmalonic acid in plasma before and after vitamin supplementation. Clin Chem 1996;42:630–6. [17] Nilsson-Ehle H, Jagenburg R, Landahl S, Lindstedt G, Swolin B, Westin J. Cyanocobalamin absorption in the elderly: results for healthy subjects and for subjects with low serum cobalamin concentration. Clin Chem 1986;32: 1368–71. [18] Hvas AM, Buhl H, Laursen NB, Hesse B, Berglund L, Nexo E. The effect of recombinant human intrinsic factor on the uptake of vitamin B12 in patients with evident vitamin B12 deficiency. Haematologica 2006;916:805–8.
Plasma holotranscobalamin compared with plasma cobalamins for assessment of vitamin B12 absorption; optimisation of a non-radioactive vitamin B12 absorption test (CobaSorb) Anne-Mette Hvas a,1 , Anne L. Morkbak b,⁎,1 , Ebba Nexo b a
b
Department of Clinical Biochemistry, Skejby Hospital, Aarhus University Hospital, 8000 Aarhus, Denmark Department of Clinical Biochemistry, Aarhus Hospital, Aarhus University Hospital, Norrebrogade 44, DK- 8000 Aarhus, Denmark Received 12 June 2006; received in revised form 7 August 2006; accepted 7 August 2006 Available online 11 August 2006
Abstract Background: A recently developed non-radioactive vitamin B12 absorption test (CobaSorb) was further explored to identify the best marker for reflection of vitamin B12 absorption and to determine the duration of the test. Methods: Seventy-eight healthy individuals (age 21–81 years) were given three oral doses of 9 μg vitamin B12 per day for 5 successive days. Non-fasting blood samples were collected on days 1 to 5 before administration of vitamin B12 and on day 8. Cobalamins and holotranscobalamin were measured. Results: Performance of the vitamin B12 absorption test was evaluated in individuals with holotranscobalamin or cobalamins below the 75% percentiles. We used a change greater than 2 × CVday-to-day in holotranscobalamin (22%) and cobalamins (12%) to indicate a change caused by absorption of vitamin B12. Among individuals with a baseline holotranscobalamin below the 75% percentile (b 75 pmol/L, n = 57), 98% had an increase in holotranscobalamin N 22% from day 1 to day 3. In contrast, only 72% of the individuals with baseline cobalamins below the 75% percentile (b 335 pmol/L, n = 57) had an increase in cobalamins N 12%. Conclusions: In healthy individuals with baseline holotranscobalamin b 75 pmol/L, vitamin B12 absorption is well reflected by an increase in holotranscobalamin after 2 days administration of oral vitamin B12. © 2006 Elsevier B.V. All rights reserved. Keywords: Absorption test; Cobalamin; Holotranscobalamin; Vitamin B12
1. Introduction Vitamin B12 is an essential micronutrient supplied by meat or dairy products. Vitamin B12 is a co-enzyme in one-carbon metabolism and essential for DNA synthesis. The clinical consequences of vitamin B12 deficiency include megaloblastic anemia and neurological disorders. In the stomach, food-bound vitamin B12 is released from dietary protein and transferred to the small intestine in complex with haptocorrin (HC). In the alkaline environment of the small intestine, HC is degraded by pancreatic enzymes. The free vitamin B12 forms a complex with intrinsic Abbreviations: HoloTC, holotranscobalamin; IF, intrinsic factor; TC, transcobalamin; HC, haptocorrin; tHcy, total homocysteine; MMA, methylmalonic acid. ⁎ Corresponding author. Tel.: +45 8949 3065; fax: +45 8949 3060. E-mail address: [email protected] (A.L. Morkbak). 1 Shared first authorship. 0009-8981/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2006.08.009
factor (IF), a protein produced in the gastric parietal cells, and the IF–vitamin B12 complex is absorbed by receptor-mediated uptake in the ileum [1–3]. If no IF is available which is seen in some patients after gastrectomy or in patients with pernicious anemia, only around 1% of the ingested vitamin B12 is taken up by passive diffusion [4]. When IF is available, approximately 75% of a single physiological dose less than 0.5 μg is absorbed and about 50% of a dose around 1.0 μg [4]. For a dose around 5 μg, the average uptake is 25%, with no further increase of the total vitamin B12 absorbed if the oral concentration is increased [4]. In the circulation, vitamin B12 is bound to two proteins, transcobalamin (TC) and HC. The TC-vitamin B12 complex (holoTC) is transported from the circulation into the cells where vitamin B12 is released after degradation of TC [5]. The physiological role of HC is still unsettled. After diagnosing vitamin B12 deficiency, the cause of deficiency has to be clarified. For many years, the Schillings test was used for
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154
this purpose [6]. In the test, labelled vitamin B12 without (Schillings test I) or with IF (Schillings test II) is administered orally. Afterwards, a large dose of vitamin B12 is injected in order to saturate endogenous vitamin B12 binders and thereby allowing the absorbed vitamin to be excreted in the urine. The amount of labelled vitamin B12 excreted in a 24-h urine sample mirrors the capacity to absorb vitamin B12. However, the cumbersome Schillings test has almost disappeared from the market for several reasons, i.e., difficulties in obtaining labelled vitamin B12 as well as native human IF for Schillings test II no longer being available. Substitution for human IF such as hog IF is difficult to obtain and is used only sparsely due to safety concerns. Thus, there is a need for alternative tests for determination of the vitamin B12 absorption. Recently, a new vitamin B12 absorption test, CobaSorb, based on oral ingestion of three 9 μg doses of non-radioactive vitamin B12 at 6-h intervals was introduced and evaluated preliminarily in healthy subjects [7] and in patients with Imerslund-Grasbeck syndrome or inherited lack of intrinsic factor [8]. After ingestion of vitamin B12, a significant increase in holoTC was seen in the healthy individuals [7,8], whereas holoTC remained unchanged in the patients [8]. In the present study, we further explored the design of CobaSorb.
151
on Biomedical Research Ethics of Aarhus County (Project Number 2005-0067).
2.2. Study design Vitamin B12 absorption was evaluated by analysis of cobalamins and holoTC in samples obtained before, during, and after intake of oral vitamin B12. The individuals took three oral doses of 9 μg cyanocobalamin (Natur-Drogeriet, Horning, Denmark) (vitamin B12) per day for 5 successive days (day 1 to 5). Vitamin B12 was administered at 6-h intervals (0800, 1400, and 2000 h), together with orange or apple juice and a snack (fruit, biscuit or bread). Non-fasting blood samples were collected on days 1 to 5 before administration of the first dose of vitamin B12 and in the morning on day 8. 2.3. Sampling and analytical methods Serum was collected for analysis of cobalamins, holoTC, and total TC (days 1 to 5 and day 8), methylmalonic acid (MMA) (day 1 and day 8) and autoantibodies against IF (day 1). EDTA blood was collected for analysis of hemoglobin and erythrocyte mean cell volume (EMV) (day 1). Lithium heparin plasma was drawn for analysis of creatinine (day 1). Heparin fluoride plasma was collected for analysis of total homocysteine (tHcy) (day 1 and day 8). Plasma for analysis of tHcy was separated from the blood cells within 2 h after sample collection and stored at − 80 °C until analysed. Cobalamins were assayed on the Advia Centaur (Bayer A/S, Germany) with a total imprecision b 10% (intra-assay imprecision b 5%). Total TC was measured by an in-house sandwich ELISA with an imprecision of 4% to 6% (intra-assay imprecision ∼ 3%) [9]. HoloTC was measured by the TC ELISA after removal of the apoTC with vitamin B12 coated beads [10]. The total imprecision was ∼ 8% [10], and the intra-
2. Materials and methods 2.1. Study population In total, 78 healthy individuals aged ≥ 18 years participated in the study. Exclusion criteria were vitamin B12 injections given within the past 5 years, use of vitamin pills containing N 1 μg vitamin B12 within the past 3 weeks, infectious disease, pregnancy or lactation, giving birth within the past 9 months, use of hormones/other systemic medication, or known chronic systemic disease. Written informed consent was obtained from all individuals, and the study protocol was approved by the Regional Committee
Table 1 Concentration of cobalamins, holotranscobalamin (holoTC), total TC, TC saturation (days 1 to 5 and day 8), serum methylmalonic acid (MMA), plasma total homocysteine (tHcy) (day 1 and 8) and IF antibodies, erythrocyte mean cell volume (EMV), hemoglobin and creatinine (day 1)
Cobalamins (pmol/L) [200–600a] HoloTC (pmol/L) [40–150b] Total TC (pmol/L) [610–1400c] TC saturation [0.05–0.20b] MMA (μmol/L) [0.08–0.28d] tHcy (μmol/L) [4.5–11.9d] IF antibodies EMV (fL) [85–100] Hemoglobin (mmol/L) Females [7.4–9.6] Males [8.4–10.8] Creatinine (μmol/L) Females [44–115] Males [62–133]
Day 1
Day 2
Day 3
Day 4
Day 5
Day 8
272 (151–528) 54 (19–133) 685 (446–1221) 0.08 (0.04–0.2) 0.1 (0.07–0.4) 7.9 (4.5–22.3) 0 91 (72–101)
298 (160–529) 78 (23–186) 651 (433–1216) 0.12 (0.04–0.3) – – – –
312 (188–648) 89 (33–186) 649 (441–1081) 0.14 (0.05–0.3) – – – –
318 (191–674) 96 (30–213) 639 (444–1073) 0.15 (0.05–0.3) – – – –
343 (201–664) 98 (29–215) 629 (478–1170) 0.16 (0.05–0.3) – – – –
309 (165–548) 62 (21–226) 654 (471–1085) 0.1 (0.04–0.3) 0.1 (0.07–0.3) 8.3 (4.2–34.5) – –
8.4 (6.9–9.5) 9.2 (7.5–10.4)
– –
– –
– –
– –
– –
68 (48–82) 81 (65–110)
– –
– –
– –
– –
– –
Median and (range) are indicated. [95% prediction intervals] are from aNexø [15]; bNexø et al. [10]; cNexø et al. [9]; and dRasmussen K. et al. [16]. The reference interval for tHcy is 4.6–8.1 μmol/L for subjects b30 years, 4.5–7.9 μmol/L for females and 6.3–11.2 μmol/L for males 30–59 years and 5.8–11.9 μmol/L for subjects N60 years.
152
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154
Table 2 Characteristics of the three individuals with serum methylmalonic acid (MMA) N0.28 μmol/L Individual
Age (years) Sex Cobalamins (pmol/L) Delta cobalamins (pmol/La) Total TC (pmol/L) HoloTC (pmol/L) Delta holoTC (pmol/La) TC saturation MMA (μmol/L) Delta MMA (μmol/Lb) tHcy (μmol/L) EMV (fL) Hemoglobin (mmol/L) Creatinine (μmol/L)
1
2
3
60 F 211 31 672 60 17 (28%) 0.09 0.29 −0.02 13.7 98 8.6 72
73 F 176 30 531 19 14 (74%) 0.04 0.30 − 0.04 14.2 91 8.9 57
53 F 172 25 662 31 13 (42%) 0.05 0.43 − 0.33 4.5 96 8.7 65
a
Increase from day 1 to day 3. Decrease in MMA from day 1 to day 8.
on the IMMULITE 2000 (Diagnostic Products Corporation, California) (total imprecision b 6%) [14]. Measurement of tHcy (day 1) failed in one individual. Hemoglobin and EMV were assessed on the Sysmex XE2100 (Sysmex, Japan) with a total imprecision of 2%. Creatinine was measured by the protein-compensated Jaffe method or the enzymatic method on the Roche Cobas Integra 800 analyzer (Roche Diagnostics, Indianapolis) with a total imprecision b 3%. 2.4. Statistical analysis The associations between age, sex and the biochemical parameters were analysed by using linear regression. Changes in means and medians were compared by paired t-test and Mann–Whitney test, respectively. To fulfil the criteria for Gaussian distribution, data were log-transformed when appropriate. P-values b 0.05 were regarded as statistically significant. All statistics were performed in Stata 8.0 (StatCorp LP, Texas).
b
3. Results assay imprecision was ~ 4% [11]. The levels of cobalamins and TC (holo and total) on days 1 to 5 and day 8 for each individual were measured in the same assay. TC saturation was calculated as holoTC/total TC. The presence of autoantibodies against IF was determined by an in-house ELISA employing immobilized recombinant IF as catching reagent and biotinylated recombinant IF as detection reagent [12]. MMA was measured by slightly modified stable-isotopedilution capillary gas chromatography mass-spectrometry (total imprecision b 8%) [13] and tHcy by the immunological method
Among the 78 individuals, 40 were males and 38 were females. The median age was 51 years, range 21–81 years, (median (range) for men was 50 years (22–81 years) and for women 52 years (21–74 years)). The biochemical characteristics are shown in Table 1. All but three individuals had normal vitamin B12 status as judged from a level of MMA within the reference interval (0.08–0.28 μmol/L [16]). Two of these individuals had low cobalamins (reference interval 200– 600 pmol/L [15]) and a low holoTC (reference interval 40– 150 pmol/L [10]), Table 2.
Fig. 1. Baseline holotranscobalamin (holoTC) versus delta holoTC and baseline cobalamins versus delta cobalamins from day 1 to day 2, day 1 to day 3 and day 1 to day 5. Broken lines indicate 2 × coefficient of variation (day-to-day) of the baseline levels of holoTC (22%) and cobalamins (12%). Individuals with a baseline holoTC ≤ 75 pmol/L or cobalamins ≤ 335 pmol/L (•); holoTC N 75 pmol/L or cobalamins N 335 pmol/L (•).
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154 Table 3 Increase in holoTC or cobalamins from day 1 to day 2, 3, 4, 5 and 8 for individuals with a holoTC ≤75 pmol/L (n = 57) or cobalamins ≤335 pmol/L (n = 57) Days
1–2 1–3 1–4 1–5 1–8
Delta holoTC (pmol/L)
Delta cobalamins (pmol/L)
Median (range) [fraction]
Median (range) [fraction]
21 (−10–73) [88%] 32 (1–61) [98%] 36 (7–84) [98%] 39 (10–106) [100%] 5 (−16–50)
30 (−39–85) [53%] 39 (−3–107) [72%] 55 (−20–141) [81%] 65 (−4–146) [86%] 35 (−46–119)
Fraction of individuals with a delta holoTC N22% or delta cobalamins N12% relative to the baseline level.
Using linear regression, we found a significant positive association between age and baseline holoTC (P = 0.03) as well as TC saturation (P = 0.04). No association was found between age and baseline total TC (P = 0.84), cobalamins (P = 0.46), MMA (P = 0.54) or tHcy (P = 0.38), respectively. Sex was not significantly associated with baseline values of holoTC (P = 0.83), cobalamins (P = 0.55), MMA (P = 0.07) or tHcy (P = 0.13). After adjusting for age and sex, a small but significant positive association was observed between baseline levels and increases in holoTC (P = 0.04) and cobalamins (P = 0.04) from days 1 to 5. Interestingly, we found a significant negative association between age and increase in holoTC (P = 0.004) from days 1 to 5. The median (range) increase in holoTC was 42 pmol/L (−2– 109 pmol/L) for individuals ≤ 50 years (n = 39) and 32 pmol/L (− 11–84) for individuals N50 years (n = 39) (P = 0.006). Age was not significantly associated with increase in cobalamins from days 1 to 5 (P = 0.06). No significant association was found between sex and increases in holoTC (P = 0.13) and cobalamins (P = 0.06). Because sex was not associated with an increase in holoTC or cobalamins and the alteration in delta holoTC with age was numerically small, we decided to analyse the total study population together. From a previous study, we know that the day-to-day variation is 11% for holoTC and 6% for cobalamins [7]. We used a change larger than 2 × coefficient of variation (day-to-day) in holoTC (22%) and cobalamins (12%) to indicate a change caused by absorption of the administered vitamin B12. As shown in Fig. 1, a substantial number of individuals with baseline levels above the 75% percentile for holoTC (75 pmol/L) or cobalamins (335 pmol/L) did not show a significant increase in holoTC (6 of 21 individuals (− 16–20%)) or cobalamins (15 of 21 individuals (− 3–12%)). Because a vitamin B12 absorption test is relevant only in individuals with borderline or low levels of holoTC or cobalamins, we analysed in detail the pattern of absorption only in those having baseline levels below the 75% percentiles. Among individuals with a baseline holoTC ≤ 75 pmol/L (b 75% percentile, n = 57), 50 (88%) had an increase in holoTC N 22% on day 2 and 56 (98%) on day 3. Among individuals with baseline cobalamins ≤335 pmol/L (b 75% percentile, n = 57), 30 (53%) had an increase in cobalamins N12% on day 2 and 41 (72%) on day 3, Table 3. The increment in both parameters was largest from day 1 to day
153
2. After day 3, the changes in holoTC and cobalamins were very modest, Table 3. All together, the median increase in holoTC from day 1 to day 3 was 55%, with a 95% prediction interval of 28–146% in individuals with a baseline holoTC ≤75 pmol/L. From day 5 to day 8, the median (range) decrease was − 30 pmol/L (− 85–15 pmol/L) for holoTC and − 35 pmol/L (− 164–39 pmol/L) for cobalamins in individuals with a baseline holoTC ≤ 75 pmol/L. On day 8, however, 2.5 days after the last dose of vitamin B12, the levels of holoTC (P b 0.0001) and cobalamins (P b 0.0001) were still higher than baseline levels. MMA was significantly lower on day 8 compared with day 1 (P = 0.001), whereas this was not the case for tHcy (P = 0.71). The group of individuals with a baseline holoTC N 75 pmol/L showed no significant decrease in MMA (P = 0.61) from day 1 to day 8. 4. Discussion In this study, changes in cobalamins and holoTC were compared regarding suitability as markers of vitamin B12 absorption. Moreover, the design of the vitamin B12 absorption test CobaSorb [7] was further explored by prolonging the duration of vitamin B12 administration. In accord with previously published results [7,8], and as expected theoretically, we found that changes in holoTC reflected vitamin B12 absorption better than changes in cobalamins. The absolute increases in cobalamins and holoTC were comparable, suggesting that the increase in cobalamins is caused by an increase in holoTC and not in holoHC. The increase in the level of cobalamins relative to baseline was low and often within the expected variation of the baseline level (Fig. 1). Therefore, we will only discuss changes in holoTC and not in cobalamins in relation to absorption of vitamin B12. We found that increased age was associated with a smaller increment in holoTC after vitamin B12 ingestion, suggesting that the ability to absorb crystalline vitamin B12 decreases with age. This was also observed in a smaller, previously published
Fig. 2. Practical use of the vitamin B12 absorption test, CobaSorb.
154
A.-M. Hvas et al. / Clinica Chimica Acta 376 (2007) 150–154
study based on CobaSorb [8]. A decline in absorption of vitamin B12 with age can be explained by a decrease in the production of IF or reduced absorptive capacity or both. In contrast to our results, a study based on the conventional Schillings test suggests no decrease in absorption of crystalline B12 with age [17]. In this study, however, only elderly persons were examined and the results were compared with previously published references intervals. Furthermore, the imprecision of the Schillings test is relatively large, and thus, an age-related decline may have been overlooked in this study. Since evaluation of vitamin B12 absorption is of interest only in patients suspected of having vitamin B12 deficiency, often based on borderline or abnormal markers for vitamin B12 status, we found it most relevant to scrutinize results from the individuals with a baseline holoTC below the 75% percentile (b75 pmol/L). These individuals showed a marked increase in holoTC one day after the start of oral vitamin B12 and a further increase the following days. Though a new steady-state was not established after 4 days of vitamin B12 administration, the increase in holoTC levelled off after 2 days. Therefore, we recommend that future testing employing the CobaSorb design include a baseline blood sample for measurement of holoTC followed by 2 days administration of three times 9 μg vitamin B12 and subsequent collection of a second sample for analysis of holoTC. Based on our data, we believe that an increase in holoTC of at least 22% supports the patient being capable of absorbing vitamin B12. Because the analytical detection limit for holoTC is 6.4 pmol/L, we additionally require an increase in holoTC ≥ 10 pmol/L. The practical use of CobaSorb is outlined in Fig. 2. A limitation of this study is that it reports results only on healthy individuals and does not include data on patients with mal-absorption of vitamin B12. It has already been documented that individuals unable to absorb vitamin B12 does not increase in holoTC one day after administration of the vitamin [8,18], but even though it is unlikely, it has to be proven that holoTC does not increase significantly in mal-absorbers after 2 days of administration of three times 9 μg vitamin B12. In conclusion, we report that in healthy individuals with baseline holoTC below the 75% percentile (b 75 pmol/L) vitamin B12 absorption is reflected by an increase in holoTC– but not cobalamins–after 2 days' administration of oral vitamin B12, and we suggest that future use of CobaSorb is done based on this design. Acknowledgement We thank Jette Fisker Petersen and Anna-Lisa Christensen at Department of Clinical Biochemistry, Aarhus Sygehus, and Lisbeth Norengaard Sørensen and Kirsten Christiansen at Department of Clinical Biochemistry, Center of Hemophilia
and Thrombosis, Skejby Hospital, Aarhus University Hospital, Denmark, for outstanding technical assistance. References [1] Moestrup SK, Verroust PJ. Megalin- and cubilin-mediated endocytosis of protein-bound vitamins, lipids, and hormones in polarized epithelia. Annu Rev Nutr 2001;21:407–28. [2] Kristiansen M, Kozyraki R, Jacobsen C, Nexo E, Verroust PJ, Moestrup SK. Molecular dissection of the intrinsic factor-vitamin B12 receptor, cubilin, discloses regions important for membrane association and ligand binding. J Biol Chem 1999;274:20540–4. [3] He Q, Madsen M, Kilkenney A, et al. Amnionless function is required for cubilin brush-border expression and intrinsic factor-cobalamin (vitamin B12) absorption in vivo. Blood 2005;106:1447–53. [4] Chanarin I. The megaloblastic anaemias. Third ed. London: Blackwell Scientific Publications; 1990. [5] Seetharam B, Yammani RR. Cobalamin transport proteins and their cellsurface receptors. Expert Rev Mol Med 2003;5:1–18. [6] Schilling RF. Intrinsic factor studies II. The effect of gastric juice on the urinary excretion of radioactivity after the oral administration of radioactive vitamin B12. J Lab Clin Med 1953;42:860–5. [7] Bor MV, Nexo E, Hvas AM. Holo-transcobalamin concentration and transcobalamin saturation reflect recent vitamin B12 absorption better than does serum vitamin B12. Clin Chem 2004;50:1043–9. [8] Bor MV, Cetin M, Aytac S, Altay C, Nexo E. Nonradioactive vitamin B12 absorption test evaluated in controls and in patients with inherited malabsorption of vitamin B12. Clin Chem 2005;51:2151–5. [9] Nexo E, Christensen AL, Petersen TE, Fedosov SN. Measurement of transcobalamin by ELISA. Clin Chem 2000;46:1643–9. [10] Nexo E, Christensen AL, Hvas AM, Petersen TE, Fedosov SN. Quantitation of holo-transcobalamin, a marker of vitamin B12 deficiency. Clin Chem 2002;48:561–2. [11] Morkbak AL, Heimdal RM, Emmens K, et al. Evaluation of the technical performance of novel holotranscobalamin (holoTC) assays in a multicenter European demonstration project. Clin Chem Lab Med 2005;43:1058–64. [12] Nexo E, Nooroya B, Hvas AM, Christensen AL, Waters H. Autoantibodies against intrinsic factor (IF) measured with an ELISA using recombinant human IF as both catching and detecting reagent. Clin Chem Lab Med 2005;43:351–6. [13] Rasmussen K. Solid-phase sample extraction for rapid determination of methylmalonic acid in serum and urine by a stable-isotope-dilution method. Clin Chem 1989;35:260–4. [14] Moller J, Ahola L, Abrahamsson L. Evaluation of the DPC IMMULITE 2000 assay for total homocysteine in plasma. Scand J Clin Lab Invest 2002;62:369–73. [15] Nexø E. Variation with age of reference values for P-cobalamins. Scand J Haematol 1983;30:430–2. [16] Rasmussen K, Moller J, Lyngbak M, Pedersen AM, Dybkjaer L. Age- and gender-specific reference intervals for total homocysteine and methylmalonic acid in plasma before and after vitamin supplementation. Clin Chem 1996;42:630–6. [17] Nilsson-Ehle H, Jagenburg R, Landahl S, Lindstedt G, Swolin B, Westin J. Cyanocobalamin absorption in the elderly: results for healthy subjects and for subjects with low serum cobalamin concentration. Clin Chem 1986;32: 1368–71. [18] Hvas AM, Buhl H, Laursen NB, Hesse B, Berglund L, Nexo E. The effect of recombinant human intrinsic factor on the uptake of vitamin B12 in patients with evident vitamin B12 deficiency. Haematologica 2006;916:805–8.