5D Library › Polypharmacy Supplement Interactions, Chronotherapy Dosing Windows and Cytochrome Enzyme Competition · 中文

Chapter 2. One Carrier, and the Losses Before It

An identity about a shared carrier

The crowded site of the previous chapter has a name and a sequence. Gunshin and colleagues cloned the proton-coupled divalent metal transporter of the small intestine and showed that it moves iron, zinc, manganese, cobalt, copper and more through the same route, which is why the minerals in a multivitamin are not independent of one another (Gunshin et al. 1997). On the clinical side, Hallberg and colleagues gave graded doses of calcium with a test meal and measured iron absorption directly, establishing the dose-response rather than the mere existence of the effect (Hallberg et al. 1991).

Evidence

Published. Gunshin and colleagues identified the divalent metal transporter and its broad substrate range (Gunshin et al. 1997). Hallberg and colleagues measured the dose-dependent inhibition of both non-haem and haem iron absorption by calcium in human subjects (Hallberg et al. 1991). Lonnerdal reviews the dietary factors that govern zinc absorption, including the molar ratio of phytate to zinc (Lonnerdal 2000).

Sweeping the calcium dose through the model turns up something the clinical literature does not usually state, because it is a property of the algebra rather than of any mineral.

Calculation

# Chapter 2. p03: What one mineral costs another at a shared carrier.
# The divalent metal carrier of the small intestine moves iron, zinc,
# magnesium and more through the same saturable route, so a large dose
# of one arrives as a competitor for the others. Here the calcium and
# magnesium doses are swept while the iron and zinc doses are held, and
# the absorbed milligrams are computed from the Michaelis fraction with
# the apparent affinity raised by the competitors present. The absorbed
# fraction is the carrier fraction times a fixed mucosal efficiency.
MASS = {"iron": 55.85, "zinc": 65.38, "calcium": 40.08,
        "magnesium": 24.305}
KM = {"iron": 0.080, "zinc": 0.150, "calcium": 5.000, "magnesium": 4.000}
EFF = {"iron": 0.18, "zinc": 0.26, "calcium": 0.30, "magnesium": 0.35}
DOSE = {"iron": 18.0, "zinc": 25.0}
SWEEP = (0.0, 100.0, 200.0, 300.0, 500.0, 700.0, 1000.0)
MG_WITH_CA = 0.8          # magnesium swept at this multiple of calcium

def mmol(name, mg):
    return mg / MASS[name]

def absorbed(name, mg, competitors):
    s = mmol(name, mg)
    extra = sum(mmol(c, d) / KM[c] for c, d in competitors.items() if d > 0.0)
    return mg * EFF[name] * s / (KM[name] * (1.0 + extra) + s)

if __name__ == "__main__":
    print("   iron {:.0f} mg and zinc {:.0f} mg held; calcium swept, with"
          .format(DOSE["iron"], DOSE["zinc"]))
    print("   magnesium at {:.1f} times the calcium dose".format(MG_WITH_CA))
    print("   calcium mg   iron mg   zinc mg   iron share   zinc share")
    base_fe = absorbed("iron", DOSE["iron"], {"zinc": DOSE["zinc"]})
    base_zn = absorbed("zinc", DOSE["zinc"], {"iron": DOSE["iron"]})
    for ca in SWEEP:
        comp_fe = {"zinc": DOSE["zinc"], "calcium": ca,
                   "magnesium": ca * MG_WITH_CA}
        comp_zn = {"iron": DOSE["iron"], "calcium": ca,
                   "magnesium": ca * MG_WITH_CA}
        fe = absorbed("iron", DOSE["iron"], comp_fe)
        zn = absorbed("zinc", DOSE["zinc"], comp_zn)
        print("   {:<13.0f}{:<10.4f}{:<10.4f}{:<13.4f}{:.4f}"
              .format(ca, fe, zn, fe / base_fe, zn / base_zn))
    print("   iron absorbed with no calcium or magnesium: {:.4f} mg"
          .format(base_fe))
    print("   zinc absorbed with no calcium or magnesium: {:.4f} mg"
          .format(base_zn))
    worst = 0.0
    for name, base, dose, other in (("iron", base_fe, DOSE["iron"], "zinc"),
                                    ("zinc", base_zn, DOSE["zinc"], "iron")):
        lo, hi = 0.0, 5000.0
        for _ in range(60):
            mid = 0.5 * (lo + hi)
            comp = {other: DOSE[other], "calcium": mid,
                    "magnesium": mid * MG_WITH_CA}
            if absorbed(name, dose, comp) > 0.5 * base:
                lo = mid
            else:
                hi = mid
        worst = max(worst, abs(0.5 * (lo + hi)))
        print("   calcium dose that halves {} absorption: {:.1f} mg"
              .format(name, 0.5 * (lo + hi)))
    print("   the two doses agree, as the algebra above requires")
    comp = {"zinc": DOSE["zinc"], "calcium": 500.0,
            "magnesium": 400.0}
    print("   at calcium 500 mg and magnesium 400 mg, the usual evening")
    print("   handful, iron keeps {:.4f} of what it would have absorbed"
          .format(absorbed("iron", DOSE["iron"], comp) / base_fe))
    print("   and the absorbed iron falls from {:.4f} to {:.4f} mg"
          .format(base_fe, absorbed("iron", DOSE["iron"], comp)))
#    iron 18 mg and zinc 25 mg held; calcium swept, with
#    magnesium at 0.8 times the calcium dose
#    calcium mg   iron mg   zinc mg   iron share   zinc share
#    0            1.7225    2.1866    1.0000       1.0000
#    100          1.4667    1.8618    0.8515       0.8515
#    200          1.2770    1.6211    0.7414       0.7414
#    300          1.1308    1.4354    0.6565       0.6565
#    500          0.9200    1.1679    0.5341       0.5341
#    700          0.7755    0.9845    0.4502       0.4502
#    1000         0.6276    0.7968    0.3644       0.3644
#    iron absorbed with no calcium or magnesium: 1.7225 mg
#    zinc absorbed with no calcium or magnesium: 2.1866 mg
#    calcium dose that halves iron absorption: 573.3 mg
#    calcium dose that halves zinc absorption: 573.3 mg
#    the two doses agree, as the algebra above requires
#    at calcium 500 mg and magnesium 400 mg, the usual evening
#    handful, iron keeps 0.5341 of what it would have absorbed
#    and the absorbed iron falls from 1.7225 to 0.9200 mg

The two right-hand columns are identical, and the comment at the head of the program explains why: once the denominator of the Michaelis expression is divided by the item’s own affinity constant, what remains is one plus the sum over every item at the site of its dose divided by its own constant, which is the same number for every occupant. When several existing occupants are present at the same time and share the same carrier, a new competitor therefore multiplies the absorption of every one of them by exactly the same factor, whatever their doses. The calcium dose that halves iron absorption and the one that halves zinc absorption are both 573.3 mg.

That is a useful thing to know and a dangerous thing to over-read. It says the shared carrier is indifferent, not that the consequences are. At the usual evening calcium and magnesium doses the absorbed iron falls from 1.7225 to 0.9200 milligrams, which keeps 0.5341 of what it would have been. Halving the absorption of a mineral nobody is short of removes nothing anyone will notice. The arithmetic is symmetric; the clinical weight is not.

Iron and zinc absorbed against the calcium dose, with the halving point marked (p03).

What never reaches the carrier at all

A second class of loss happens earlier, in the lumen, and has nothing to do with competition for a carrier. A mineral can be bound by something it was eaten with, or the chemistry it needs can be absent. Hurrell, Reddy and Cook measured the inhibition of non-haem iron absorption by tea, coffee and other polyphenol-containing drinks and found effects large enough to matter at ordinary serving sizes (Hurrell et al. 1999). Lonnerdal’s review treats the phytate-to-zinc molar ratio as the organising variable for zinc, which is why the ratio and not the phytate mass is what the program computes (Lonnerdal 2000).

Calculation

# Chapter 2. p04: Losses that happen before the carrier is reached. Some
# of a dose never arrives at a transporter at all, because it has been
# bound in the gut lumen or because the chemistry it needs has been
# removed. Three such losses are costed here: phytate binding of zinc,
# read off the molar ratio of phytate to zinc that the nutrition
# literature uses; polyphenol binding of non-haem iron, as happens when
# a mineral is taken with tea or coffee; and the loss of iron solubility
# when an antacid or a large carbonate dose raises the stomach's pH.
# The multipliers are illustrative fits to the published shapes.
ZN_MASS, FE_MASS, PHYTATE_MASS = 65.38, 55.85, 660.0
ZN_DOSE, FE_DOSE = 25.0, 18.0
MEALS = (
    # name, phytate mg, polyphenol mg, stomach pH
    ("water, empty stomach",       0.0,    0.0, 1.6),
    ("eggs and spinach",         180.0,   40.0, 1.8),
    ("oat porridge",             900.0,   30.0, 2.0),
    ("wholemeal bread and tea", 1100.0,  320.0, 2.1),
    ("with a proton pump pill",    0.0,    0.0, 5.0),
    ("carbonate antacid too",      0.0,    0.0, 6.2),
)
PH_HALF, PH_HILL = 3.4, 2.4        # iron solubility against pH

def zinc_factor(phytate_mg):
    ratio = (phytate_mg / PHYTATE_MASS) / (ZN_DOSE / ZN_MASS)
    return 1.0 / (1.0 + 0.55 * ratio)

def polyphenol_factor(mg):
    return 1.0 / (1.0 + mg / 120.0)

def ph_factor(ph):
    return 1.0 / (1.0 + (ph / PH_HALF) ** PH_HILL)

if __name__ == "__main__":
    print("   setting                   phytate  zinc    iron    iron")
    print("   ratio to zinc             factor   polyph  pH      total")
    rows = []
    for name, phy, pol, ph in MEALS:
        ratio = (phy / PHYTATE_MASS) / (ZN_DOSE / ZN_MASS)
        zf = zinc_factor(phy)
        pf = polyphenol_factor(pol)
        hf = ph_factor(ph)
        rows.append((name, ratio, zf, pf, hf, pf * hf))
        print("   {:<26}{:<9.3f}{:<8.4f}{:<8.4f}{:<8.4f}"
              .format(name, ratio, zf, pf, hf))
    print("   iron surviving both the polyphenol and the pH loss:")
    for name, ratio, zf, pf, hf, tot in rows:
        print("   {:<26}{:.4f}".format(name, tot))
    best = max(rows, key=lambda r: r[5])
    worst = min(rows, key=lambda r: r[5])
    print("   best setting for iron: {} at {:.4f}".format(best[0], best[5]))
    print("   worst: {} at {:.4f}".format(worst[0], worst[5]))
    print("   ratio between them: {:.2f}".format(best[5] / worst[5]))
    print("   zinc: best {:.4f}, worst {:.4f}, ratio {:.2f}"
          .format(max(r[2] for r in rows), min(r[2] for r in rows),
                  max(r[2] for r in rows) / min(r[2] for r in rows)))
    print("   milligrams of iron reaching the carrier from an {:.0f} mg"
          " dose:".format(FE_DOSE))
    for name, ratio, zf, pf, hf, tot in rows:
        print("   {:<26}{:.4f}".format(name, FE_DOSE * tot))
    print("   the pH term alone costs {:.4f} of the dose on a proton"
          " pump pill".format(1.0 - ph_factor(5.0) / ph_factor(1.6)))
    lo, hi = 1.0, 8.0
    for _ in range(60):
        mid = 0.5 * (lo + hi)
        if ph_factor(mid) / ph_factor(1.6) > 0.5:
            lo = mid
        else:
            hi = mid
    print("   stomach pH at which iron solubility halves: {:.3f}"
          .format(0.5 * (lo + hi)))
#    setting                   phytate  zinc    iron    iron
#    ratio to zinc             factor   polyph  pH      total
#    water, empty stomach      0.000    1.0000  1.0000  0.8592
#    eggs and spinach          0.713    0.7182  0.7500  0.8215
#    oat porridge              3.566    0.3377  0.8000  0.7813
#    wholemeal bread and tea   4.359    0.2944  0.2727  0.7607
#    with a proton pump pill   0.000    1.0000  1.0000  0.2838
#    carbonate antacid too     0.000    1.0000  1.0000  0.1913
#    iron surviving both the polyphenol and the pH loss:
#    water, empty stomach      0.8592
#    eggs and spinach          0.6161
#    oat porridge              0.6251
#    wholemeal bread and tea   0.2075
#    with a proton pump pill   0.2838
#    carbonate antacid too     0.1913
#    best setting for iron: water, empty stomach at 0.8592
#    worst: carbonate antacid too at 0.1913
#    ratio between them: 4.49
#    zinc: best 1.0000, worst 0.2944, ratio 3.40
#    milligrams of iron reaching the carrier from an 18 mg dose:
#    water, empty stomach      15.4665
#    eggs and spinach          11.0899
#    oat porridge              11.2513
#    wholemeal bread and tea   3.7342
#    with a proton pump pill   5.1088
#    carbonate antacid too     3.4426
#    the pH term alone costs 0.6697 of the dose on a proton pump pill
#    stomach pH at which iron solubility halves: 3.826

Three settings deserve attention. Wholemeal bread with tea gives a phytate-to-zinc molar ratio of 4.359, which leaves zinc at 0.2944, and the polyphenol term leaves iron at 0.2727, so iron survives at 0.2075 and only 3.7342 milligrams of an 18 milligram dose reach the carrier. A proton pump pill, which changes no food at all, costs iron 0.6697 through the acidity term alone, and with a carbonate antacid as well iron survives at 0.1913. Between the best and worst settings in the table iron differs by 4.49 times and zinc by 3.40 times. The pH at which iron solubility halves is 3.826, which is inside the range an acid-suppressing medicine routinely produces.

Note

Why this matters more than the carrier does. The losses in this section are larger than the competition losses of Chapter 1 and are easier to fix, because they depend on what else is in the stomach and when, not on how much of each mineral was bought. A person on long-term acid suppression who takes iron is in the worst cell of this table every day, and no adjustment of the iron dose addresses it. That is a conversation for a doctor, not for a schedule.

Hypothesis · not established

The author’s interpretation (not yet tested): for a person on acid suppression, the luminal loss dominates every carrier-level interaction, so separating their mineral doses in time will produce no measurable benefit while the acid suppression continues. What would show it wrong: in a crossover study of people on a proton pump inhibitor, if separating the minerals in time raised absorbed iron as much as it does in people not on one, the luminal term would not be dominant.

Summary

The shared carrier treats its occupants alike: a new competitor multiplies every existing occupant’s absorption by the same factor, so the calcium dose that halves iron and the one that halves zinc are both 573.3 mg, and at the usual evening doses iron keeps 0.5341. The losses before the carrier are larger: iron survives at 0.2075 with bread and tea and at 0.1913 with an antacid, acidity alone costs 0.6697 on an acid-suppressing medicine, and solubility halves at pH 3.826. Chapter 3 multiplies the two kinds of loss together.

References

Gunshin, Hiromi, Bryan Mackenzie, Urs V. Berger, et al. 1997. “Cloning and Characterization of a Mammalian Proton-Coupled Metal-Ion Transporter.” Nature 388 (6641): 482 to 488.
Hallberg, Leif, Mats Brune, Monica Erlandsson, Ann-Sofie Sandberg, and Lena Rossander-Hulthen. 1991. “Calcium: Effect of Different Amounts on Nonheme- and Heme-Iron Absorption in Humans.” The American Journal of Clinical Nutrition 53 (1): 112 to 119.
Hurrell, Richard F., Manju Reddy, and James D. Cook. 1999. “Inhibition of Non-Haem Iron Absorption in Man by Polyphenolic-Containing Beverages.” British Journal of Nutrition 81 (4): 289 to 295.
Lonnerdal, Bo. 2000. “Dietary Factors Influencing Zinc Absorption.” The Journal of Nutrition 130 (5): 1378S to 1383S.

Fan, Y. (2026). Polypharmacy Supplement Interactions, Chronotherapy Dosing Windows and Cytochrome Enzyme Competition: Michaelis-Menten Inhibition Matrices, Chelation Absorption Losses and Circadian Receptor Sensitivity [复合补剂相互作用、时间疗法给药窗口与代谢酶竞争] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090250 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).