Elevated cholesterol, particularly low density lipoprotein (LDL) cholesterol, is a well defined risk factor for the development of atherosclerosis (NCEP, 2002). Atherosclerotic lesions are thought to form in response to endothelial damage caused by excess LDL (Steinberg, 2005). Most cases of cardiovascular disease (CVD), the leading cause of mortality and disability in the developed countries, are ascribed to pathology of atherosclerosis. Strategies to lower plasma LDL cholesterol levels are therefore crucial in the prevention and treatment of CVD (O'Keefe et al., 2009). Conversely to plasma LDL levels, elevated high density lipoproteins (HDL) are protective against the development of CVD. Low density lipoproteins are responsible for the transport of cholesterol and fatty acids from the liver to the various tissues of the body. In contrast HDL is responsible for the majority of reverse cholesterol transport (RCT), in which cholesterol from the peripheral cells are returned to the liver for excretion in the bile (Joy and Hegele, 2008). It is this major role that HDL plays in RCT that is attributed with HDL's anti-atherogenic effects. Therefore strategies which elevate HDL levels are also beneficial in the prevention and treatment of CVD.
Plasma cholesterol can be synthesized by the body (hepatic or extra-hepatic sources) or absorbed from the intestines (dietary or biliary sources). Statins, which are a family of HMG-COA reductase inhibitors, reduce cholesterol synthesis and are very effective and can achieve plasma cholesterol lowering of up to 55% and have been shown to reduce CVD risk by one-third (Stein, 2002). Statin therapy has also been shown to increase plasma HDL levels from 2-16%, it is not know if this increase is clinically significant to the efficacy of statin therapy as it is difficult to separate this small and variable increase from the benefit of the significant LDL lowering (Hou and Goldberg, 2009). Regardless of the statins' efficacy of return LDL cholesterol levels to their desired range, a significant proportion of statin users continue to have adverse events (Alsheikh-Ali et al., 2007). Therapies which reduce intestinal cholesterol or bile acid absorption are also available. Plant sterol/stanols, dietary fibre, bile acid sequestrants have been shown to be effective, with LDL cholesterol reductions of 10-15%, 8.5-13% and 5-30% respectively (Hou and Goldberg, 2009). The drug Ezetimibe, which binds to Niemann-Pick C1 Like 1 (NPC1L1) protein a crucial mediator of cholesterol absorption (Chang and Chang, 2008), has also been demonstrated to lower LDL cholesterol levels by 16-19% (Pandor et al., 2009). The wide ranges of cholesterol lowering seen in each treatment are likely the product of genetic factors which vary the rates of cholesterol synthesis and absorption, as well as modulate the effectiveness of each intervention. Because of this genetic heterogeneity of cholesterol synthesis and absorption, being able to effectively measure these parameters and how they respond to different dietary, pharmaceutical or lifestyle interventions is paramount to understanding interventions effect on cholesterol metabolism.
Methods to assess cholesterol absorption
A direct method of measuring cholesterol absorption refers to measuring cholesterol flux from the intestines to the lymph. It is a direct method because it does not use a faecal or plasma level of cholesterol to infer absorption (Gibson, 1984). The first direct method of measuring cholesterol absorption required a duodenal cannula for infusion and cannulisation of the mesenteric or thoracic lymph ducts to sample lymph (Pouteau et al., 2003a). This method has been successfully used to access cholesterol absorption in rats, dogs and rabbits. The use of radioactively labelled cholesterol in the duodenal infusion allows for the distinction between exogenous and billiary cholesterol and can be used to calculate the percent absorption of exogenous cholesterol. An alternative direct method for use in humans was introduced by Grundy and Mok (1977). The intestinal perfusion method (sometimes refered to as Method VII) requires intubation with a three-lumen tube. Liquid formula and ß-sitosterol (though to be a non-absorbable cholesterol degradation marker at the time) is infused into the duodenum at the Ampulla of Vater though the first tube, and intestinal content is aspirated from the 2nd and 3rd tube, 10 and 100 cm distal to the first, respectively. Analysis of the aspirate from allows for the calculation of net hourly cholesterol absorption across the 100 cm long segment of intestine. Discriminating between exogenous and endogenous sources of cholesterol can be done by infusing radioactive cholesterol. The use of ß-sitosterol as a non-absorbable marker in this method may not have been the most appropriate choice given what is known about plant sterols and inhibition of cholesterol absorption at even very low levels (Calpe-Berdiel et al., 2009; Ostlund et al., 2002a; Ostlund et al., 2002b). Plant sterols complete with cholesterol for absorption into gut enterocytes, but are then exported back into the intestines by ABCG5 and ABCG8 transporters. This inhibition was seen by Grundy and Mok, who noticed that even small increases in ß-sitosterol concentration caused noticeable reductions in measured cholesterol absorption (Grundy and Mok, 1977). The use of an inhibitor of cholesterol absorption in a method to access cholesterol absorption obviously leads to underestimation of actual cholesterol absorption capacity, and this must be taken into account when comparing absorption values among different methods. Another downside to the direct measurements techniques is their invasiveness, however they are the only methods which yield direct measures of total, exogenous and endogenous cholesterol absorption across a section of the intestine (Gibson, 1984).
Cholesterol balance methods
Cholesterol balance methods use the differences between dietary cholesterol and faecal cholesterol, excluding cholesterol of endogenous origins, to estimate cholesterol absorption (Matthan and Lichtenstein, 2004). Radioactive labelled cholesterol is used to distinguish endogenous from exogenous cholesterol and faecal and plasma levels of cholesterol are measured. There are 4 main balance methods, following the nomenclatures from Grundy and Arhens (1969), they are Method I, II, III and V.
Method I consists of a single dose of [14C] or [3H] radio-isotope labelled cholesterol administered intravenously (Grundy and Ahrens, 1969; Quintao et al., 1971). This radio-isotope labels endogenous cholesterol and its subsequent steroid products. Using the formula: daily exogenous absorbed cholesterol (mg/day) = daily cholesterol intake (mg/day) – daily unabsorbed dietary cholesterol (mg/day), and daily unabsorbed cholesterol (mg/day) = faecal neutral steroids per day (mg/day) - endogenous faecal neutral steroids per day (mg/day). Endogenous faecal neutral steroids are calculated by dividing the total radioactivity (dpm/day) in the total faecal neutral steroids by the specific activity (dpm/mg) of the plasma cholesterol 1-2 days prior (depending on gastrointestinal transit time). Method II requires continuous oral labelling with radioactive cholesterol (Grundy and Ahrens, 1969; Quintao et al., 1971). The radio-labelled cholesterol is generally incorporated into a liquid diet to insure a precise daily intake of radiosterol for many weeks. Cholesterol absorption is calculated using the formula Z = X + Y, and Z * SA(Z)= X *SA(X) + Y * SA(Y), where Z = total faecal neutral steroids (mg/day), SA(Z) = specific activity of total faecal neutral steroid (dpm/mg), X = faecal neutral steroids of endogenous origin (mg/day), SA(X) = specific activity of plasma cholesterol (dpm/mg), Y= daily unabsorbed dietary cholesterol (mg/day), and SA(Y) = specific activity of dietary cholesterol (dpm/mg). The equations are solved for Y, and absorption measurements can be done at any time after 4 days of continuous radioisotope feeding, a isotopic steady state is not required. It has been shown that Method II provides more consistent data on cholesterol absorption than, and is therefore recommended over, Method I (Gibson, 1984; Matthan and Lichtenstein, 2004).
Method III was introduced by Wilson and Lindsey (1965) and requires an isotopic steady state. Continuous feeding with radioisotope labelled cholesterol for greater than 100 days is required to reach an isotopic steady state. When an isotopic steady state is reached daily cholesterol turnover is calculated, and daily absorbed dietary cholesterol is calculated by the formula: Dietary cholesterol absorption (mg/day) = daily cholesterol turnover (mg/day) * (specific activity of plasma cholesterol/specific activity of dietary cholesterol). Failure to reach an isotopic steady state will yield an underestimate of actual dietary cholesterol absorption (Gibson, 1984). The long period required to reach an isotopic steady state and difficulty determining when this state has been reached make Method III a difficult and unreliable method. Method III was shown to give lower values of cholesterol absorption than both Method I and II in direct comparison (Quintao et al., 1971).
Method V (Grundy et al., 1971) is a combination of methods I and II. Subjects are continuously fed isotopically labelled cholesterol as in Method I and then given a single bolus of different isotopically labelled cholesterol as in Method II. The equations for Method II are used, with only the modification of the X *SA(X) term which is replaced by [(R)(SA(X oral)) / (SA(X intravenous))], where R= daily faecal neutral steroid excretion of intravenous cholesterol (mg/day), SA(X oral) = plasma specific activity of oral isotope taken 1 day before R (dpm/mg), SA(X intravenous) = plasma specific activity of intravenous isotope taken 1 day before R. Method V was designed for situation when rates of cholesterol synthesis are unusually high (due to surgical or pharmacological interference with enterohepatic circulation), which can lead to endogenous cholesterol being secreted into the intestines prior to isotopic equilibration(Gibson, 1984). In this circumstance (very high synthesis) Method I would underestimate and Method II would overestimate cholesterol absorption.
Methods I, II and V all require a marker for gastrointestinal transit time, and for cholesterol degradation in the gut. Chromic oxide is often used as a marker of gastrointestinal transit time. ß-sitosterol has been used as a control for cholesterol degradation, however it must again be mentioned that if ß-sitosterol is used as a marker then the cholesterol absorption values obtained are likely lower than the actual values. However, in these methods if the degradation of cholesterol in the gut is not accounted for, the calculated cholesterol absorption will be higher than the actual values (Quintao et al., 1971).
Isotope ratio methods
Isotope ratio methods measure only the percentage cholesterol absorption (Pouteau et al., 2003a), unlike balance methods, and require accurate record of dietary cholesterol intake to estimate the mass of exogenous cholesterol absorbed from the percentage (Matthan and Lichtenstein, 2004).
Method IV (also called faecal isotope ratio method), first introduced by Borgstrom (1969) calculates cholesterol absorption as the percentage of a single oral dose of radio-labelled cholesterol not recovered in the faeces. A single 1 μCi radio-labelled dose of both cholesterol and ß-sitosterol is administered orally and feces are collected for seven to eight days. The ratio of labelled cholesterol to labelled ß-sitosterol in a sample of the pooled faeces samples is compared to the ratio that was given orally and cholesterol absorption is calculated using the formula: Cholesterol Absorption (%) = 1 –((Faecal Cholesterol (dpm) /Faecal ß-sitosterol (dpm)) * (Oral Cholesterol (dpm)/Oral ß-sitosterol (dpm))) *100. This method has been modified to allow a single faecal sampling, Sodhi et al. (1974) administered a faecal flow marker, such as chromic oxide or carmine red, with the original test dose, and calculated the faecal isotopic ratio at the peak level of the marker in the faeces rather than a sample of the pooled faeces. This single faecal sampling modification of Method IV has been compared to the original Method IV and Method I in baboons and it yielded consistently higher absorption levels, suggesting that this modification may weaken the accuracy (Mott et al., 1980). Method IV, despite the long period of faecal collection required, uses a far smaller dose of radioactivity than the balance methods, and is relatively straightforward and simplistic in execution (Gibson, 1984). It does still require ß-sitosterol as a marker of cholesterol degradation in the gut, so it shares the drawbacks previously discussed.
The plasma isotope ratio method (IRM or Method VI), was first introduced by Zilversmit (1972) in rats and then subsequently carried out in humans (Samuel et al., 1978). The IRM involves simultaneous oral and intravenous administration of C14 and H3 radio-labelled cholesterol and requires only a single blood sample 3-4 days afterwards. The methods principles are based on the measurement of drug absorption used in pharmacology. If the absorption of cholesterol was 100% than the specific activity curve of both radio labelled cholesterols would be the same, if absorption was zero, none of the orally administered cholesterol, and therefore zero radioactivity from the oral radioisotope would appear in the plasma. Since cholesterol absorption falls between zero and 100%, the ratio of the two plasma radioactivities, after normalization for dose, are used to calculate absorption using the formula: Cholesterol Absorption (%) = 100 * (% oral dose in plasma/ % IV dose in plasma), where % oral dose and % IV dose in plasma are the percentage of IV and oral tracer in the plasma sample, respectively (Samuel et al., 1982). The ratio is calculated 3-4 days after radioisotope administration because the specific activity time curves of both radioisotopes must be parallel. This does not occur in human until 3-4 days due to a delay in the appearance of the oral radioisotope in the blood related to the mechanisms of cholesterol absorption (Gibson, 1984). The IRM advantages over previous methods include: 1)only a single blood sample is required, 2) a low dose of radioisotope is used, 3)it does not require faecal collection 4) it does not required markers, such as ß-sitosterol, to correct for faecal losses. The IRM lends it's self to repeated use because of its short duration and low level of labelling, this allows for investigation of variations in cholesterol absorption under different experimental parameters in the same individual in a comparatively short time period. The IRM method has been validated in human numerous times against method IV under different conditions yielding similar results(Samuel et al., 1978; Samuel et al., 1982).
A third isotope ratio method, Method VIII was introduced by Crouse and Grundy (1978), it shares similarities to Method IV, average cholesterol absorption is calculated using the ratio administered cholesterol to ß-sitosterol measured in the faeces, but differs in the method of isotope administration. C14 Cholesterol and H3 ß-sitosterol are administered orally three times daily for 10 days, and faeces can be collected from day 4-10 (Gibson, 1984; Matthan and Lichtenstein, 2004). Following day 3 of isotope administration the ratio of isotopes in the faeces becomes essentially constant and cholesterol absorption is calculated by the formula:
Absorption (%)= 100* ((Faecal cholesterol (dpm)/Faecal sitosterol (dpm))/1-dietary cholesterol (dpm)/ dietary sitosterol (dpm))
This method requires only the ratio of radioactivity in a single faecal sample to be measured; faecal mass need not be calculated (Crouse and Grundy, 1978). Since ß-sitosterol is also administered with the labelled cholesterol it may underestimate actual cholesterol absorption.
Plasma cholesterol specific radioactivity following the consumption of a test meal has also been investigated a measure of cholesterol absorption. Lin et al. (2005) measured cholesterol absorption in 11 individuals with Smith-Lemli-Opitz syndrome (SLOS), a cholesterol synthesis disorder, and compared plasma cholesterol apecific activity with cholesterol absorption measured by Method IV. They sampled blood 24, and 48 hours following radioisotope enriched tests meals, and calculated the specific radioactivity of the cholesterol in the plasma (dpm/mg cholesterol). These values correlated significantly with cholesterol absorption calculated by Method IV (r=0.594, p=0.009, and r=0.474, p=0.047 for 24 and 48 hours respectively). While this method cannot calculate cholesterol mass or percent absorption, it could allow for investigation of relative changes in cholesterol absorption within an individual across different conditions. It requires only a single blood sample, no stool collection, and no use of a cholesterol degradation marker, such as β-sitosterol. This method is very similar, to the single isotope tracer method that will be discussed in the stable isotope methods section (Wang et al., 2004). This relationship between plasma radioactivity at 24 hours and cholesterol absorption requires further validation, especially in healthy individuals.
The use of radio-isotopes has been invaluable to the investigation of cholesterol absorption, however the advent of stable isotope laboratory techniques, especially developments in isotope ratio mass spectrometry has allowed radio-isotopes such as C14 and H3 tracers to be replaced with safer 13C, 2H and 18O stable isotope labelled tracers (Pouteau et al., 2003a). This has reduced the difficulty related to containment, handling, disposal and overall safety associated with radioisotopes and allowed for investigation of cholesterol absorption in certain populations (children, pregnant and lactating mothers) which had previously been impossible do to the ethical considerations around radioisotope administration.
Stable isotope methods for cholesterol absorption
Cholesterol labelled with stable isotopes has been shown to have identical kinetics as radio-labelled cholesterol. This has led to the development of stable isotope techniques to investigate cholesterol absorption. Lutjohann et al. (1993) introduced a stable isotope version of Crouse and Grundy 's (1978) Method VIII discussed above. Deuterated cholesterol and sitostanol were used in place of C14 cholesterol and H3 ß-sitosterol, and quantified using gas-liquid chromatography -selected ion monitoring. Cholesterol absorption was calculated as in Method VIII. The stable isotope method was twice compared to Method VIII, in six monkeys, yielding similar results. The stable isotope method produced and absorption range of 49-73% (mean of 60%), with coefficient of variation ranging from 3.9%-15.1% (mean 7.1%). The radioisotope produced a range of 51-69% (mean 61%) with coefficient of variation ranging from 1.9-13.6% (mean 5.1%) (Lutjohann et al., 1993). This stable isotope Method VIII was determined to be as effective as the radio-isotope Method VIII, without the risk of radioactive exposure to subjects and research staff. This method uses sitostanol, rather than β-sitosterol, as a marker for faecal losses of cholesterol. This, however, does not remedy the problem other methods suffer from when using β-sitosterol, as sitostanol is also an inhibitor of cholesterol absorption (Gylling et al., 1997; Miettinen et al., 1995).
The plasma isotope ratio method (Method VI) (Samuel et al., 1982; Zilversmit, 1972) was also adapted to stable isotopes. Bosner et al. (1993) used 2H labelled oral and 13C labelled IV cholesterol to calculate the plasma stable isotope ratio and cholesterol absorption percentage. This method uses gas-chromatography – mass spectrometry with select ion monitoring (GC/MS-SIM) or GS/MS- chemical ionization mode (CI) to determine isotopic enrichment. Bosner et al. (1999) further modified the Method VI to a single isotope dual tracer method, using oral [2H]5 and IV [2H]6 cholesterol. Isotope detection in the plasma cholesterol is done by GC/MS- selected mass monitoring. Jones et al. (2000) were the first to use isotope ratio mass spectrometry to determine isotopic enrichment using the plasma isotope ratio method. Oral 13C and IV 3H cholesterol were administered to 15 hypercholesterolemic men, followed by blood sampling 2-3 days post administration. Free cholesterol from red blood cells was purified by thin layer chromatography and subsequently combusted to yield carbon dioxide and water. The CO2 was then measured for 13C enrichment against the international standard Pee Dee Belemetite (PDB) on an isotope ratio mass spectrometer (IRMS). The water was reduced to hydrogen gas via zinc reduction and 3H enrichment was measured against Standard Mean Ocean Water (SMOW) international standard by IRMS. The ratio of plasma enrichment of 13C to 3H cholesterol on day 3 after tracer administration was used to calculate cholesterol absorption. Recently, continuous flow gas chromatograph pyrolysis IRMS (GC/P/IRMS) systems and 18O or 3H cholesterol with gas chromatograph combustion IRMS (GC/C/IRMS) systems and 13C cholesterol have been used in calculating cholesterol absorption using IRM (Method VI)(Gremaud et al., 2001; Pouteau et al., 2003b). The use of IRMS vs GC/MS-SIM has vastly increased the precision of this method.
The two single stable isotope methods for accessing cholesterol absorption were introduced by Ostlund et al. (1999) and Wang et al. (Wang et al., 2004). Ostlund et al. (1999) administered deuterated cholesterol to volunteers and measured the average oral cholesterol tracer in plasma ((mmol deuterated cholesterol/mol natural cholesterol) in blood samples taken 4 and 5 days post tracer administration using GC/MS. Wang et al. (2004) administered 13C cholesterol orally followed by blood sampling at 24, 48, 72 and 96 hours. 13C enrichment in plasma free cholesterol was measured using GC/C/IRMS. Average 13C enrichment from 24-96 hours and area under the curve (24-96 hours) of 13C enrichment were compared the cholesterol absorption percentage measured using the plasma dual stable isotope ratio method (Bosner et al., 1993) in 2 studies. The average and area under the curve of 13C enrichment in plasma free cholesterol correlated with cholesterol absorption percentage measured by stable isotope method VI (r values ranging from r=0.81, p=0.0001 to r=0.88, p=0.0001)(Wang et al., 2004). Both of these single isotope methods are used to compare treatment effects, such as pharmaceutical or dietary interventions and their effects relative to control, on cholesterol absorption.
Absorption surrogate method
The use of serum plant sterol levels to predict cholesterol absorption was first developed by Tilvis and Miettinen (1986). They showed that serum levels of ß-sitosterol and campesterol, when normalized for total serum cholesterol, correlated positively with cholesterol absorption as measured by Method VIII of Crouse and Grundy (1978). This measurement involves lipid extraction from a single blood sample followed by GC or HPLC to quantify serum plant sterol and cholesterol levels. The use of campesterol or ß-sitosterol to cholesterol ratio has since been used numerous times as a measure of cholesterol absorption (Gylling et al., 2007; Hallikainen et al., 2006; Matthan et al., 2009; Nissinen et al., 2008; Simonen et al., 2008). Nissinen et al. (2008) showed that ß-sitosterol to cholesterol ratio correlated better with cholesterol absorption than campesterol to cholesterol ratio, as measured by method VIII, across three diets varying in both cholesterol and fat levels in 29 healthy male volunteers. It is important that when using serum plant sterol as surrogates for cholesterol absorption it is imperative that factors which are known to change serum plant sterol levels, such as the dietary intake of plant sterols (Chan et al., 2006), should be controlled for so as not to perceive a change in cholesterol absorption which may not exist (Vanstone and Jones, 2004). Serum plant sterol levels can be variable within and across different population (Chan et al., 2006), they can also be severely elevated in certain individuals due to genetic disorders (Berge et al., 2000; Lee et al., 2001). It is important that individuals with these genetic disorders are not included in studies using this method of estimating cholesterol absorption as serum plant sterol levels do not reflect cholesterol absorption in these individuals. Recently it has also been shown that serum plant sterol concentrations do not reflect cholesterol absorption in individuals with Smith-Lemli-Opitz syndrome (Merkens et al., 2009), therefore the use of serum plant sterols as surrogates for cholesterol absorption should be verified prior to its use in a particular populations. The use of plant sterol surrogates does benefit from relative speed and simplicity compared to other previous discussed methods (Matthan and Lichtenstein, 2004; Pouteau et al., 2003a), it is the only estimate of cholesterol absorption which can be done in large scale studies.
Methods to access cholesterol synthesis
The primary pharmaceutical intervention used to lower elevated cholesterol levels are statins, which are inhibitors of HMG-CoA reductase, a key enzyme in cholesterol synthesis (Stein, 2002). Cholesterol synthesis contributes more to circulating cholesterol levels than cholesterol absorption (Dietschy, 1984; Pouteau et al., 2003a), and maintains cholesterol levels during fasting, therefore accurate assessment of cholesterol synthesis is essential to the field of cholesterol research.
Cholesterol synthesis can be estimated when intake of dietary cholesterol and excretion of total cholesterol is known during a metabolic steady state (Grundy and Ahrens, 1969). The criteria of this metabolic steady state are: constant plasma cholesterol and faecal cholesterol excretion levels during a period of constant weight. In this steady state, cholesterol synthesis is the difference between cholesterol excretion (faecal neutral sterols and bile acids) and intake. Dietary cholesterol intake must be accurately measured, and faeces must be collected, for a given period of time. Faecal flow must also me monitored to assure collection of faeces from the appropriate time period. Faecal neutral sterols and bile acids are measured in the faeces, typically by GC-MS, and cholesterol synthesis for a given period of time is calculated (Jones et al., 1998; Kempen et al., 1988). Although this method is the gold standard for calculating cholesterol synthesis, it is vulnerable to errors in the estimation of both cholesterol intake and excretion which can potentially cause significant error. The balance method does determine the actual mass of cholesterol synthesized during a given period, it requires a metabolic steady state, accurate measurement of cholesterol intake and total faecal collection for the period of interest. Therefore this method is not suited for larger trials.
Fractional conversion of squalene
Daily cholesterol synthesis rate can be estimated by the fraction of infused radiolabelled melavonic acid converted to cholesterol (Liu et al., 1975; McNamara et al., 1977). [14C] mevalonic acid and [3H] cholesterol are administered intravenously, this infusion of labelled mevalonic acid rapidly labels the plasma squalene pool, reaching a maximum enrichment at ~100 minutes. Cholesterol synthesis is estimated by measuring squalene synthesis. Squalene synthesis is calculated by the percentage of melavonate dose converted to cholesterol, divided by the area under the curve of pasma squalene specific activity. This method assumes that plasma squalene synthesis is equivalent to cholesterol synthesis. The cholesterol synthesis rate estimated using this method has been shown to agree with cholesterol synthesis calculated by the balance method within 8% (Liu et al., 1975). The benefits of this method is it requires only 1 hour of a participant's time, and can be repeated every 3 weeks (McNamara et al., 1977), it does however require the administration of intavenous radioisotopes.
Cholesterol synthesis precursors
The concentrations of plasma precursors along the synthesis pathway of cholesterol have been used to as an indirect qualitative measure of cholesterol synthesis. Squalene (Miettinen, 1982), mevalonic acid (Parker et al., 1982), lanosterol, desmosterol and lathosterol have all been used a surrogates for cholesterol synthesis (Matthan et al., 2000; Miettinen et al., 1990). These precursors have has been shown to fluctuate with diurnal cholesterol synthesis patterns as well as in conditions in which cholesterol synthesis is elevated or reduced (Miettinen, 1982). These precursors correlate more closely with measured cholesterol synthesis when they are normalized for plasma cholesterol level, then as the absolute amount of precursor, and are normally expressed in mmol/mol cholesterol (Miettinen et al., 1990; Nissinen et al., 2008). When using cholesterol precursors as surrogates for cholesterol synthesis, dietary intakes of each precursor should be controlled for, this is particularly important for squalene, which is abundant in olive oil and is at least partially absorbed into the blood (Ostlund et al., 2002b). The most consistent surrogate for cholesterol synthesis has been lathosterol (Kempen et al., 1988; Nissinen et al., 2008). Since cholesterol synthesis surrogates require only one blood sample they are ideal for estimating cholesterol synthesis in very large studies or epidemiological trials.
Mass Isotopomer Distribution Analysis (MIDA)
Mass isotopomer distribution analysis (MIDA) is a technique that can be used to measure the synthesis of biological polymers in vivo (Hellerstein and Neese, 1999). The technique uses the relative abundance (pattern or distribution) of polymer species which differ only in mass (mass isotopomers) produced during the administration of stable isotope labelled precursors. The distribution of the polymer species produced is compared to the theoretical distributions predicted by binomial and polynomial expansion. Using these theoretical distributions, parameters such a fractional synthesis rate can be calculated using combinatorial probability model. Since cholesterol is synthesized from subunits of acetyl-Coa, fractional synthesis of cholesterol can be calculated during the infusion of 13C labelled acetate (Neese et al., 1993). This method is very invasive, requiring a 24-hr intravenous infusion and serial blood sampling from a indwelling catheter (Di Buono et al., 2000). The cholesterol in the blood samples is measured by GC-MS to determine the distribution pattern of isotopomers from which the rate of synthesis is obtained. The data analysis relies on more complex mathematical modelling than other methods for estimating cholesterol synthesis.
This method is based on the tritiated water uptake method by Dietchy and Spady (1984) used originally in animals. Deuterium incorporation method uses water as a tracer to determine the synthesis of free cholesterol (FC). The fractional synthesis rate (in pools/day) of free cholesterol is calculated from the rate of incorporation of deuterated water into de novo synthesized plasma or erythrocyte cholesterol. Orally administered deuterated water equilibrates with the body's water and NADPH. Body water and NADPH account are the precursors for 22 of the 46 hydrogen in synthesized cholesterol (Jones, 1990). Deuterium enrichment of the precursor pool, plasma water, and in erythrocyte or plasma cholesterol are measured by IRMS. Fractional synthesis rate of free cholesterol (FSR-FC) is calculated using the following formula:
FSR-FC (pools/day) = ( δ cholesterol ‰/ δ plasma water ‰ * 0.478)
δ refers to the change in deuterium enrichment over 24 hours and 0.478 is the ratio of cholesterol from body water and NADPH to total hydrogen in a cholesterol molecule, or the ratio of hydrogen which could be enriched by oral D2O administration (Jones et al., 1993). From the FSR-FC the ASR -FC (g/day) can be calculated by multiplying the FSR-FC by the M1 pool size (determined by Goodman's equation) and 0.33 the proportion of FC in total cholesterol. The ASR-FC is approximately the daily production of newly synthesized cholesterol. The three main assumptions required for deuterium incorporation are: 1)That the fraction of hydrogen derived from plasma water (22/46, or 0.478) is constant in denovo synthesized cholesterol. 2) That denovo synthesized free cholesterol rapidly exchanges between the site of synthesis and the major free cholesterol (plasma compartment) pool. Within this pool cholesterol migrates rapidly between cellular membranes and lipoproteins as well as between different classes of lipoproteins. 3) Deuterium uptake into the free cholesterol of the major (plasma) pool of cholesterol reflects synthesis in this pool and that synthesis of cholesterol in the major pool provides a reasonable measurement of total cholesterol synthesis as most sterol synthesis occurs in the intestines and liver which contribute to the plasma cholesterol pool. It must be acknowledged that the major (plasma) pool of cholesterol is at equilibrium with two other slow turnover pools outside the plasma, and that the slow inter-pool cholesterol exchange would cause insignificant entry of labelled free cholesterol into the central pool within a 24 hour time period (Dietschy and Spady, 1984; Jones, 1990). Although it has been shown that these assumptions are not perfect, the cholesterol synthesis estimates yielded by deuterium incorporation have been sensitive enough to show differences in cholesterol synthesis due to genetic factors, and dietary and pharmaceutical interventions. It has also been shown to correlate well with cholesterol synthesis measured by the balance method (Jones et al., 1998), MIDA(Di Buono et al., 2000) and cholesterol synthesis surrogate levels (Matthan et al., 2000).
The methods used to quantify cholesterol absorption and synthesis reviewed in this article have yielded invaluable information, as well as provided effective means of measuring the experimental effects of various interventions, on cholesterol homeostasis. The methods have evolved substantially over the years, from radio-isotopes to stable isotopes, and from highly invasive to less invasive procedures. However, advancement in the quantification of cholesterol homeostasis must still be continued, with a goal of finding more accurate methods to measure cholesterol synthesis and absorption, perhaps simultaneously, something which until now is still unavailable. The advantages and drawbacks, as well as the type of information yielded by each method for measuring cholesterol absorption and synthesis must be weighed carefully when selecting an appropriate method. The cost and available technical expertise and facilities will limit which methods are available to each investigator. However, it is imperative that the assumptions and limitations of each method are checked to insure that its use is valid for each particular experimental question it is used to answer.