VLDL Triglyceride Production and Clearance

From Hepatic Fatty-Acid Supply to VLDL Output

Plasma triglyceride concentration reflects both how much triglyceride the liver secretes and how rapidly triglyceride is removed from circulating lipoproteins. A useful mechanistic model must therefore connect hepatic substrate supply to secretion, then connect secretion to plasma triglyceride dynamics. This tutorial builds that connection with a compact mass balance and a saturable clearance term.

The model distinguishes fatty acids taken up by the liver from de novo lipogenesis (DNL), the synthesis of fatty acids within the liver. These are separate inputs to hepatic triglyceride synthesis. In particular, DNL is not counted as part of nonesterified fatty-acid uptake. The earlier material on Adipose Lipolysis and Free-Fatty-Acid Feedback can inform how circulating fatty-acid supply changes; here, that supply enters the liver model as an input.

Definition: Let $U_N$ be hepatic uptake of nonesterified fatty acids (NEFA), in moles of fatty-acid equivalents per unit time, and let $D$ be hepatic DNL, in the same units. Let $a_N$ and $a_D$ be the fractions of those respective inputs routed toward the hepatic triglyceride pool. Let $H$ be the amount of triglyceride in that pool. Since one triglyceride molecule contains three fatty-acid chains, the routed fatty-acid input in triglyceride equivalents is $$ J_{\mathrm{in}}=\frac{a_N U_N+a_D D}{3}. $$ If $k_{\mathrm{ox}}$ represents first-order removal from the pool by oxidation and other non-VLDL disposal, and $k_{\mathrm{sec}}$ represents first-order VLDL secretion, the hepatic balance is $$ \frac{dH}{dt}=J_{\mathrm{in}}-(k_{\mathrm{ox}}+k_{\mathrm{sec}})H, \qquad S=k_{\mathrm{sec}}H, $$ where $S$ is VLDL-triglyceride secretion.

The division by three is a stoichiometric conversion: the inputs are fatty-acid equivalents, whereas $H$ and $S$ are measured as triglyceride amount and triglyceride amount per time. The routing fractions account for the fact that not every fatty acid taken up or synthesized is directed to triglyceride. The model assumes these fractions and the rate constants are fixed over the interval being considered.

This is a deliberately aggregated liver model. It treats hepatic triglyceride as one well-mixed pool, assumes input and removal rates are adequately represented by the stated expressions, and omits changes in pool composition. It also omits other possible sources of hepatic fatty acids, such as dietary or remnant-derived inputs. Those sources would need to be added explicitly if they matter to the question being studied.

Worked Example: Calculating Hepatic Secretion From Two Substrate Sources

Suppose a hypothetical liver takes up $12$ mmol fatty-acid equivalents/day from NEFA and produces $3$ mmol fatty-acid equivalents/day by DNL. Let $a_N=0.65$, $a_D=0.80$, $k_{\mathrm{ox}}=0.04$ day$^{-1}$, and $k_{\mathrm{sec}}=0.06$ day$^{-1}$. Calculate the steady hepatic triglyceride pool and VLDL-triglyceride secretion.

First calculate the routed inputs. The NEFA contribution is $0.65(12)=7.8$ mmol fatty-acid equivalents/day, and the DNL contribution is $0.80(3)=2.4$ mmol fatty-acid equivalents/day. Thus, $$ J_{\mathrm{in}}=\frac{7.8+2.4}{3}=3.4\ \text{mmol triglyceride/day}. $$

At steady state, $dH/dt=0$, so $J_{\mathrm{in}}=(k_{\mathrm{ox}}+k_{\mathrm{sec}})H$. Therefore, $$ H_{\mathrm{ss}}=\frac{3.4}{0.04+0.06}=34\ \text{mmol triglyceride}. $$

The secretion rate is $S=k_{\mathrm{sec}}H_{\mathrm{ss}}$, giving $$ S=0.06(34)=2.04\ \text{mmol triglyceride/day}. $$ The remaining $0.04(34)=1.36$ mmol/day is removed through the model’s non-VLDL disposal pathway. Together, these two removal fluxes equal the $3.4$ mmol/day input, as required by mass balance.

Plasma Triglyceride Dynamics and LPL-Mediated Removal

Let $C$ denote plasma triglyceride concentration. For a fixed plasma volume $V_p$, secretion $S$ corresponds to an appearance rate $q=S/V_p$ in concentration per unit time. A simple representation of lipoprotein-lipase (LPL)-mediated triglyceride removal is a saturable rate:

Model: With $V_{\max}$ expressed in concentration per time and $K_m$ in concentration, the plasma balance is $$ \frac{dC}{dt}=q-\frac{V_{\max}C}{K_m+C}. $$ Here $q$ is the rate of triglyceride appearance in plasma, and the second term is the rate of triglyceride removal attributed to LPL-mediated processing. The model assumes constant $q$, $V_{\max}$, $K_m$, and plasma volume for the period analyzed. It tracks triglyceride mass in plasma; it does not assert that LPL hydrolysis removes the entire lipoprotein particle from circulation.

The saturable term increases with $C$, but approaches $V_{\max}$ as $C$ becomes large. At concentrations much lower than $K_m$, it is approximately first-order: $V_{\max}C/(K_m+C)\approx (V_{\max}/K_m)C$. At concentrations comparable to or greater than $K_m$, that approximation can underestimate the degree of saturation and misrepresent the concentration response.

For constant positive input, a finite steady state exists only if the input is less than the maximum removal capacity. Setting $dC/dt=0$ and solving gives the exact steady-state relationship.

q=\frac{V_{\max}C_{\mathrm{ss}}}{K_m+C_{\mathrm{ss}}} \quad\Longrightarrow\quad C_{\mathrm{ss}}=\frac{qK_m}{V_{\max}-q}, \qquad 0<q<V_{\max}.

The condition follows directly from the algebra: $q(K_m+C_{\mathrm{ss}})=V_{\max}C_{\mathrm{ss}}$, so $qK_m=(V_{\max}-q)C_{\mathrm{ss}}$. A positive finite solution therefore requires $V_{\max}-q>0$. If $q=V_{\max}$, the removal rate approaches the input only asymptotically as concentration grows; there is no finite steady concentration. If $q>V_{\max}$, removal cannot balance input at any concentration, so this model predicts continuing accumulation rather than a finite steady state.

Worked Example: Finding a Plasma Triglyceride Steady State

Suppose a hypothetical system has a constant plasma appearance rate $q=0.68$ mmol/L/day, $V_{\max}=1.20$ mmol/L/day, and $K_m=0.20$ mmol/L. Find the predicted steady concentration and verify it by substitution.

Because $0.68<1.20$, a finite positive steady state is possible. Substitution into the steady-state formula gives $$ C_{\mathrm{ss}}=\frac{(0.68)(0.20)}{1.20-0.68} =\frac{0.136}{0.52} =0.262\ \text{mmol/L}. $$

At this concentration, the removal rate is $$ \frac{(1.20)(0.262)}{0.20+0.262} \approx \frac{0.3144}{0.462} \approx 0.680\ \text{mmol/L/day}. $$ The removal and appearance rates agree to rounding, confirming the balance. This calculation is illustrative; it is not a measurement from a clinical study.

Derivation, Stability, and a Linked Perturbation

The steady-state formula describes where the model’s net rate of change is zero. It is also important to check whether a small displacement from that state produces a restoring tendency. Define the net rate as $f(C)=q-V_{\max}C/(K_m+C)$. For positive $K_m$ and $V_{\max}$, $$ \frac{df}{dC}=-\frac{V_{\max}K_m}{(K_m+C)^2}<0. $$ Consequently, increasing $C$ above a steady state makes the net rate negative, while decreasing $C$ below it makes the net rate positive. The steady state is locally stable under the model’s constant-input assumptions.

The same relationship shows why concentrations can respond disproportionately when input approaches the clearance capacity. As $q$ approaches $V_{\max}$ from below, the denominator $V_{\max}-q$ becomes small, so $C_{\mathrm{ss}}$ rises sharply. This is a feature of the specified saturable-clearance model, not a general claim that all biological systems have this exact response.

Worked Example: Increasing DNL and Propagating Its Effect

Use the hypothetical hepatic parameters from the first worked example, and suppose the plasma volume is $3.0$ L. Increase DNL from $3$ to $6$ mmol fatty-acid equivalents/day while keeping all other inputs and parameters fixed. Calculate the new hepatic secretion rate and plasma steady state.

The routed input is now $$ J_{\mathrm{in}}=\frac{(0.65)(12)+(0.80)(6)}{3} =\frac{7.8+4.8}{3} =4.2\ \text{mmol triglyceride/day}. $$ The hepatic steady pool is $H_{\mathrm{ss}}=4.2/(0.04+0.06)=42$ mmol triglyceride, so secretion is $S=0.06(42)=2.52$ mmol/day. The plasma appearance rate is therefore $q=2.52/3.0=0.84$ mmol/L/day.

Using $V_{\max}=1.20$ mmol/L/day and $K_m=0.20$ mmol/L, the new plasma steady state is $$ C_{\mathrm{ss}}=\frac{(0.84)(0.20)}{1.20-0.84} =\frac{0.168}{0.36} =0.467\ \text{mmol/L}. $$ This is higher than the $0.262$ mmol/L baseline. DNL doubled, but total routed input did not double because NEFA uptake remained unchanged. The concentration increase is larger than the increase in secretion because the higher input moves the system closer to its assumed maximum clearance rate.

Assumptions, Diagnostics, and Alternatives

Interpret predictions in light of the model’s assumptions. Inputs and parameters are treated as constant, plasma is represented by one well-mixed compartment, hepatic secretion responds through one triglyceride pool, and LPL-mediated removal is summarized by one saturable rate. The equations do not separately represent time-varying feeding, changes in plasma volume, multiple lipoprotein populations, or the subsequent fate of particles after triglyceride hydrolysis. The model is most defensible when those omitted processes do not dominate the time course or question of interest.

Useful checks include confirming that units match in each balance, that $a_N$ and $a_D$ lie between zero and one, and that calculated fluxes are nonnegative. For a proposed steady state, verify $q<V_{\max}$ and substitute the predicted concentration back into the removal term. In time-course data, persistent increases in concentration despite a predicted steady state may indicate changing input, an incorrect clearance capacity, or a missing compartment or process. A single steady concentration generally cannot identify both $V_{\max}$ and $K_m$: different combinations can produce the same removal rate at that concentration. Independent information or observations across a range of concentrations are needed to distinguish them reliably.

A linear first-order clearance model, $dC/dt=q-kC$, is a simpler alternative when data support proportional removal over the observed concentration range. Its steady state is $q/k$, and it does not impose a finite maximum removal rate. The saturable model is preferable when clearance capacity is a relevant mechanistic hypothesis, but its extra parameters require enough informative data. Neither formulation should be selected solely because it gives a convenient fit.

Common pitfall: It is incorrect to treat every fatty acid taken up or synthesized by the liver as secreted VLDL triglyceride. That reasoning ignores routing, competing disposal, and the hepatic pool balance. The correction is to convert fatty-acid equivalents to triglyceride equivalents, account for the fraction routed to the pool, and calculate secretion from the pool’s secretion term. A second warning is that $q\geq V_{\max}$ does not imply an extremely high finite steady state: under this model, no finite steady state exists.

Check Your Understanding

Answer each question using the model equations above. Solutions follow the questions.

If $U_N=9$ and $D=3$ mmol fatty-acid equivalents/day, with $a_N=0.50$ and $a_D=1.00$, what is $J_{\mathrm{in}}$ in mmol triglyceride/day?
For $J_{\mathrm{in}}=2.0$ mmol triglyceride/day, $k_{\mathrm{ox}}=0.03$ day$^{-1}$, and $k_{\mathrm{sec}}=0.07$ day$^{-1}$, calculate $H_{\mathrm{ss}}$ and $S$.
For $q=0.50$ mmol/L/day, $V_{\max}=1.00$ mmol/L/day, and $K_m=0.25$ mmol/L, calculate $C_{\mathrm{ss}}$.
What does the plasma model predict if $q=1.10$ mmol/L/day and $V_{\max}=1.00$ mmol/L/day? Explain why.

Solutions. For question 1, $J_{\mathrm{in}}=[(0.50)(9)+(1.00)(3)]/3=(4.5+3)/3=2.5$ mmol triglyceride/day. For question 2, $H_{\mathrm{ss}}=2.0/(0.03+0.07)=20$ mmol triglyceride, and $S=0.07(20)=1.4$ mmol/day. For question 3, $C_{\mathrm{ss} }=(0.50)(0.25)/(1.00-0.50)=0.125/0.50=0.250$ mmol/L. For question 4, there is no finite steady state: the input exceeds maximum removal capacity, so $q>V_{\max}$ and the model predicts continuing accumulation.

VLDL Triglyceride Production and Clearance

What You'll Learn