clays and binders, thiols and sulfur

DMSA and ALA transporters

DMSA and ALA bind mercury. How are they transported in our bodies? .Candidates include OAT1, MRP2, and the Na+ multi-vitamin transporter.

0. Background Lipoic acid: Na+ Multi vitamin transporter

The SMVT was considered a transporter of LA due to competition with substrates biotin and pantothenic acid. The need for LA transport is fuzzy as this review claims synthesiis in the mitochondria. Indeed, LA is due to the fact that LA inhibited the uptake of the other two SMVT substrates, biotin and pantothenic acid, in concentration-dependent manner. Indeed, R-LA serves as a cofactor in the pyruvate dehydrogenase and α-ketoglutarate dehydrogenase complexes, enzymes that catalyze central redox reactions related to energy production in the cell as well as in oxidative decarboxylations of α-keto acids and amino acids [1]

A collage of Slc5A6, also known as the sodium dependent multi-vitamin transporter. Upper and middle left: data from Protein Atlas. Upper right structures of substrates for SMVT Bottom a brain capillary and structures of Hg-lipoic acid conjugates from a previous post.

Protein Atlas does not have data for immunolocalization of NMVT in the choroid plexus. mRNA levels of this part of the brain are very high. The immunocytochemistry from the cerebral cortex show blood vessel staning (top middle). The three recognized substrates of SMVT share commonality of structures. We still do not know how lipoic acid-Hg chelates get out of the endothelial cells.

0. Background DMSA: via OAT1/ and MRP2

The export of DMSA-Hg has been addressed in another way in a previous post. Kobayashi and coworkers used a cell culture system expressing various transporters. Instead of using DMSA-Hg they used a DMSA radioactive isotope of technium. Thesource of data is Protein Atlas MRP2. Slc13A3, aka the high-affinity sodium-dicarboxylate cotransporter recognizes substrates with 4-6 carbon atoms: succinate, alpha-ketoglutarate and N-acetylaspartate. The stoichiometry is probably 3 Na(+) for 1 divalent succinate

From ProteinAtlas and ref [2] MCME Megalin/Cubilin mediated endocytosis Upper left, OAT3; upper right, MRP2, lower right,

The reuptake of the Tc-DMSA complex in the proximal tubule is not explored in this post.

1. Does oral chelator get absorbed well?

We need basolaeral and apical transporters for these chelators. This section addresses bio availability.


About 66% of orally administrated [14C] α-Lipoic was found in blood compared to compared to the intraveneous route. . An alternative evaluation by comparison of [14C]material excreted into the urine yielded 93% [14C]absorption. These authors also ligated the small intestinal tract to test the hypothesis that different regions differed in efficiency. [3] The abstract was truncated and effort was not expended tracking down the publication. Also note that the was racemic. Racemic means it contains both stereo isomers. The R (for right handed) form of α-Lipoic may be transported differently than the left handed S form (S for sinister).

Structure of R-α-lipoic acid (A) and S-α-lipoic acid (B). Chiral center shown with asterisk (*).[4]

These are the structures. Does the oral vs intracuodenal rout influence the amount in the plasma?

acokinetic ParametersOralIntraduodenal
Cmax (µg/mL) in plasma2.8 ± 0.52.4 ± 0.6 *14.7 ± 1.711.5 ± 1.7 *
Tmax (min) time to Cmax2.0 ± 0.02.0 ± 0.05.0 ± 0.04.3 ± 1.5
AUC (µg·min/mL)48.1 ± 15.638.8 ± 13.2 *154.2 ± 11.3116.5 ± 4.4 *
Table 3 from ref [4] Pharmacokinetic parameters are shown as mean ± standard deviation (n = 4). RLA, R-α-lipoic acid; SLA, S-α-lipoic acid; Cmax, maximum plasma concentration; Tmax, time of maximum plasma concentration; AUC, area under the curve plasma concentration versus time curve from time 0 to the last; *, probability (p) <0.01 compared with RLA. Statistical analysis was performed by using the paired-t test.

Other Uchida et al publication notes

  • Cmax was one fifth by the oral route as it was for intravenous.
  • One the other hand, the half life in the blood via the oral route (about 26 minutes both isomers) was twice the intra venous route (about 11 minutes both isomers) .
  • Upon intra-portal injection, the R form was significantly (p<0.01) more bioavailable than the S-form 7.5 ± 6.1 41.0 ± 5.1, AUC (µg·min/mL), respectively.
  • After 60 minutes at stomach pH 1.2, only 12% of both isoforms of α-lipoic acid remained. [4]


This is a cut and paste summary of DMSA mixed disulfide abstract [5]

By 14 hr after DMSA administration (10 mg/kg), only 2.5% of the administered DMSA is excreted in the urine as unaltered DMSA and 18.1% of the dose is found in the urine as altered forms of DMSA. Most altered DMSA in the urine is in the form of a mixed disulfide. It consists of DMSA in disulfide linkages with two molecules of L-cysteine. One molecule of cysteine is attached to each of the sulfur atoms of DMSA. The remaining 10% of the altered DMSA was in the form of cyclic disulfides of DMSA. So far, the mixed disulfide has been found in human but not in rabbit, mouse, or rat urine.[5]

Pb poisoned children urinating DMSA-conjugates and Pb

In a 1995 Asiedu and coauthors measured DMSA in the plasma of volunteers after oral ingestion. [6] This was a followup to better understand the pharmo-kinetics in children with Pb poisoning.

Figure 1, Asiedu 1995 [6] Shown here are two representative subjects. Arrow mark time points in which the altered DMSA tracts the Pb in the urine. Note the unaltered DMSA is much less at these time points. The authors used these data to claim that the modified DSMA was the active chelator.

Three healthy adults were recruited to participate additional experiments to better understand the pharmokinetics.

DMSA is absorbed via cholesterol chylomicrons

In two examples shown in their Figure 2 Cholestyramine was used to bind cholesterol and block it’s absorption.

  • Administration at 4, 8, and 12 hr abolished all evidence of DMSA reentry via the enterohepatic circulation.
  • The AUC(5-15 hr) was decreased by 20% (t = 13.643, p <0.001; paired t-test).
  • Cmax and t1/2 were not significantly affected.
  • With just DMSA, 18.3 ± 4.9% of the administered dose was recovered in urine in 15 hr. The range was 11.0-26.3%.
  • The unaltered form, i.e. not conjugated to cysteine,was 11 to 25% across all subjects. The authors did not make it clear if these included the children with lead poisoning and/or adults used to better understand the pharmokinetics. The conclusion was that enterohepatic circulation of the cysteinated and/or the glucaronicated forms were reabsorbed.

The next experiment was used to use the antibiotic Neomycin to test the hypothesis that microflora plays a role in circulation of DMSA [6] One or more metabolites of DMSA undergoes entero-hepatic circulation. , Neomycin decreased AUC and abolished the postprandial peaks of DMSA in every subject. Neomycin may have bound DMSA. The conclusion was that gut microflora hydrolyze poorly absorb able polar conjugates of DMSA, generating more readily absorbed unconjugated form(s),

[6] Figure 2. Semilog plots of plasma total meso-2,3-dimercaptosuccinic acid (DMSA) concentration versus time in two typical normal adults (panels A and B) following the administration of a single dose of 10 mg/kg DMSA at time zero. When DMSA was given alone, the plasma DMSA concentration peaked at 2.5 hr and then fell gradually; postprandial peaks were evident. When DMSA was given with cholestyramine (CME), the plasma DMSA concentration again peaked at 2.5 hrbut fell gradually without any evidence of postprandial reentry. When DMSA was given with neomycin, the post prandial peak in DMSA in the plasma was abolished. [7] Twenty years later these data were fit to a mathematic model.

This post is going to skip the evaluation of a rather lengthy discussion of Uchida and coauthors on mixed disulfides and so on. van Eijkeren were able to fit an updated model to their data 20 years later . [7] What does this model include? The answer is not much. These new authors were still not sure whether DMSA chelated Pb on its on or if it were a pro-drug.

A simple model of DMSA absorption and elimination [7] that does not take into account that DMSA may be a pro-drug. Structures of Pb disodium EDTA and cystine from PubChem are included.

When one compares something close to the structure of the DMSA mixed disulfde and PbEDTA, the model of Uchida seems to be very plausible If we mix DMSA with cystine at around pH 8-9 we might form the mixed disulfide anyway. Come to think of it, this is what happens in the duodenum. With all of this enterohepatic circulation, the mixed disulfide might be bio available.

2. absorbed intact without degradation?

α-Lipoic seems to be degraded just a bit in the stomach. [4] DMSA seems to be prone to mixed disulfide formation with cysteine. This could be an issue and/or asset with cysteine containing peptides in the small intestine.

3. once inside the enterocyte, does it get released to plasma?

If this question is asked from the point of view of a renal tubule epithelial cell, the exit if via the multi-resistance protein 2 in the introduction. A 2010 study demonstrated basolateral to blood efflux of glucornide conjugates greatest in the duodenum followed by the jejunum, ileum, and colon. [8]

4. where does it go next? to an organ? or just float in plasma for a while?

Two monoesters of meso-2,3-dimercaptosuccinic acid (DMSA), monoisoamyl meso-2,3-dimercaptosuccinate (Mi-ADMS) and mono-n-hexyl meso-2,3-dimercaptosuccinate (Mn-HDMS) were compared to DMSA in their efficiency to mobilize 203Hg in mercury-laden suckling rats. Seven-day-old pups were given 203Hg (18.5 kBq) with a dose of 0.5 mg Hg/kg/day as HgCl2 for five days. Seven days after the beginning of Hg loading a ten-day oral treatment with DMSA, Mi-ADMS, or Mn-HDMS was administered at a dose of 0.25 mmol/kg/day. At the end of experiment, radioactivity was measured in the whole body, liver, both kidneys, and brain.

data from the Kostial abstract [9] in table form

The esters are obviously a lot better at chelating Hg. Would they be better at chelating Pb? Similar information for lipoic acid was not found. If ALA gets into tissues via the Na+ dependent multi-viatmin transporter, it may go everywhere. ALA and DHLA appear to hitch a ride on albumin with molar ratios of 10 ± 2, 7.9 ± 2. [10] Both were able to replace octanoic acid from albumin.

5. does it cross BBB?

The results of Kostial 1995 suggests that Mia-DMSA gets past the BBB better than DMSA. OR it finds another less direct means of lowerint the Hg content. [9] A similar study dosed rats with sodium meta-arsenite (25 ppm in drinking water) for 12 weeks) . A detox program was performed as indicated. Neurological assessments and activities of enzymes associated with oxidative stress reduction were also performed [11] The goal of this experiment was to determine if lipid encapsulated nanoparticles outperformed the bulke Mia-DMSA.

Fig. 8. Estimation of arsenic levels in blood, brain, kidney, and liver of rats following exposure to arsenic and treatment with MiADMSA and nano-MiADMSA. Metals levels in the blood: μg/dl; brain: ng X10/g; liver: μg/g;kidney: μg/g; Values are mean ± SEM, n = 5. *p < 0.05 when arsenic exposed animals
compared to naive. †p < 0.05 when bulk and nanoencapsulated MiADMSAcompared with arsenic exposed.

In this particular example it does not seem like that much arsenic is getting deposited in the brain in the first place even though the mia-DMSA seems to be improving things.

Lipoic acid, does it go to the brain in large amounts?

Crossing the blood brain barrier seems to be much more dubious. One the other hand, there is the role of lipoic acid in the mitochondria off BBB cells. The following are bullet points from an abstract [12]

  • Brain and blood samples were assayed after LA dosing at 50 mg/kg.
  • LA as determined via liquid chromatography tandem mass spectrometry after residual blood was removed from the brain..
  • LA in control rats fluctuated between 0.005 and 0.267 microM in blood and 0-0.024 microM in brain.
  • After dosing brain LA levels ranged from 0.0009 to 0.0072 microM.
  • The in vitro and in vivo LA brain:blood partition ratios were 0.1 and -0.01, respectively.
  • These . [12]

Another study of allergic encephalomyelitis examined monocyte invasion of endothelial monocytes as well as changes in the cytosketon in rsponse to reactive oxygen species generators. LA prevented these deleterious changes in the cytoskeleton. [13]

6. if floats in plasma, what makes it leave in the kidney? just competitive chance,

One would think that these conjugates get filtered at the level of the glomerulus. A 1989 publication made the case for glomerular filtration and peri-tubular reabsorption of Technium-99m-dimercaptosuccinic acid. [14] From an energetics stand point, this makes way too much sense to pursue further. [14] The PubMed searches on lipoic acid and glomerulus hits the same snag as in the blood brain barrier: lipoic acid alters the structure of the glomerulus too. Podocytes can be reduced in number in diabetgic nephropathy. [15] Diabetes was induced by STZ injection in rats. Podocyte number at 6 weeks after STZ decreased. LA returned the numbers to the control level. Volume per podocyte was also increased bySTZ and normalized by LA. It becomes difficult to sort the roles of LA as an antioxidant and as a mitochondrial cofactor.

7. Does it bind heavy metal in the plasma?

These are some considerations that are hard to find addressed on PubMed

  1. Albumin, and probably other plasma proteins, also binds Hg and transition metals. What are the relative numbers of these sites and the number of sites that can reasonably be achieved on non-toxic levels of ALA and DMSA?
  2. Heavy metal binding to albumin means that it could be degraded via salvage pathways. How do the kinetics of eliminating these heavy metal bound peptides compare to chlelating agents like DMSA?
  3. Can ALA bind Hg with the same affinity when ALA is bound to albumin?

8. DMSA or DMPS Hg into the proximal renal tubule cell?

It does not seem like it which is a bummer, Alpha-1-microglobulin binds to

Chelators can bind to proteins that are reabsorbed [16]

Observation: 99mTc-labeled dimercaptosuccinic acid (99mTc-DMSA) accumulates in the kidney cortex and is widely used for imaging of the renal parenchyma. Despite its extensive clinical use, the mechanism for renal targeting of the tracer is unresolved. These authors tested the hypothesis that the megalin and cubilin albumin transpot ensemble had something to do with this. The authors used wild type and knock out mice to test the hypothesis that technium radioactivity would end up in the urine in transport deficient mice. 99mTc-MAF3 was used fo comparison.

Control or megalin/cubilin-deficient mice were injected intravenously with 0.5 MBq of 99mTc-DMSA or 99mTc-mercaptoacetyltriglycine (MAG3). Whole-body scintigrams and the activity in plasma, urine, and the kidneys were examined 6 h after injection.

What protein is 99mTc-DMSA binding to?

Fractionation studies

Urine from megalin/cubilin-deficient mice injected with 99mTc-DMSA or control mice (n = 3 in each group) was pooled, and 800 μL were applied to molecular weight cut off centrifuge filters. Only proteins smaller than the cutoff (Amicon; Millipore) with a molecular weight cutoffs of 100, 30, 10, 3, and less than 3 kDa, after centrifugation at 4,000g.[16]

molecular weight (kDa)Control miceMegalin/cubilin-deficient micerepresentative protein
30–100943albumin, microglobulin
ABLE 1 Ultrafiltration of Urine from Mice Injected with 99mTc-DMSA. Values are from mixed urines (n = 3) and expressed as percentage of injected dose.
  • 4A. Urine proteins were resolved by size via SDS PAGE The proteins were transferred to a membran that was scanned for radioactivity. Note that wildtype mice lacked radioactive proteins because they were absorbed. The megalin/cubilin knockout mice had one primary radioactive protein band.
  • 4B When similar gels were stained for proteins, it became obvious that the knockout mice had a much greater protein load in their urine.
  • 4C, an immunoblot proving that the 25 kDa protein is what they think it was.
Panel 6Mechanism for renal uptake of 99mTc-DMSA. (A) α1-microglobulin–bound 99mTc-DMSA is freely filtered by glomeruli and accumulates in renal proximal tubules by endocytosis mediated by multiligand-binding megalin/cubilin receptors (B). (C) Free 99mTc-DMSA and trace amounts of α1-microglobulin–bound 99mTc-DMSA are excreted in urine. Consequently, megalin/cubilin dysfunction leads to abolishment of renal uptake and increased urinary excretion of α1-microglobulin–bound 99mTc-DMSA.

The authors argued for lysosomal degradation of 99mTc-DMSA bound microglobulin because its uptake was impaired in CIC5 knockout mice. CLCN5 is a gene that codes for an antiporter that helps control the pH of hte lysosomes in renal epithelial cells. Is excretion by proximal tubular cells, is not disturbed by defective endocytosis.

Does some of this DMSA-Hg complex leave via liver?

From just reading the abstract, rats were given these dithiols i.v. DMPA was detected in the bile in the unaltered and DTT reducible form. DTT, dithiothreitol is a common laboratory reducing agent used to reduce disulfide bonds. These disulfide bonds can include protein, GSH and cysteine bonds.[17]

compoundbililaryunalteredDTT reducable
N-(2,3-Dimercaptopropyl) phthalamidic acid (DMPA)72%50%50%
meso-dimercaptosuccinic acid (DMSAnot foundnot found
2,3-dimercapto-1-propanesulfonic acid (DMPS) found
  • DMPA (0.10 mmol/kg), given to rats 3 days after exposure to Cd, elicited within 30 min a 20-fold increase in biliary Cd excretion.
  • The increase of biliary Cd by DMPA was dose-related and not due to an increase of bile flow rate.
  • DMSA and DMPS did not significantly affect the biliary excretion of Cd.
  • Incubation of DMPA or DMSA with Cd-saturated metallothionein (MT) resulted in the removal of Cd from MT. DMPA was more active than DMSA in this respect. .

The working model is that DMPA enters the cells, steals Cd from metallothionein, and gets excreted in the biles. [17]

9. Or is the heavy metal already in the kidney via some other natural binder such as albumin, glutathione, etc.

This brings us back to the Bridges model of GSH/ albumin Hg detoxification model of the Bridges lab. So heavy metals are binding to proteins that might elicit aggregation and target the proteins for autophagy. A recent review on lysosome depletion in renal injury had some interesting things to say about heavy metals [19] :

  1. HgCl2 impairs lysosome function in tubular cells. We can speculate that Hg binds the active site thiol of cathepsins.
  2. Pb inhibits lysosome acidification by binding to two V-ATPase subunits H+ pump subunits. Blockage of autophagic flux and lysosome membrane permeabilization LMP contributes to caspase-3-mediated apoptosis in lead-treated tubular cells.
  3. In cadmium nephrotoxicity, autophagy is protective under light loads. Heavy loads disrupt lysosome stability and lysosome flux protects against tubular injury under low cadmium stress. However, high cadmium stress disrupts lysosome stability and impairs autophagic flux.. Just scanning abstracts separate from the review [19] inhibition of lyosome-autophagosome fusion has something to do wit it.
A summary figure from the Chen review modified to account for actions of Hg, Pb, and Cd

Heavy metals recieved scant mention in this review on the many ways that sabotaging endosomal sorting complex required for transport (ESCRT) can cause renal disease. [19] Proteins filtered through the glomerular filtration barrier can be reabsorbed by tubule epithelial cells (TECs), followed by digestion and reuse.

10. DMSA & DMPS keep Hg and other heavy metals, already in the kidney so they can be transported to the lumen.???

Mercury seems to harm these transporters more than anything else. DMSA and DMPS have different transporters to get them into the proximal renal tubule cells so depending on the person, one may be better than the other. Combination may be helpful for some. I am leery about DMPS though

On the other hand, the Chen review suggests a new way of looking at things. Heavy metals get stuck in the tubule epitehlial cells when heavy metals much up lysosome processing of proteins that are binding the heavy metals.

Concluding remarks


  1. Zehnpfennig B, Wiriyasermkul P, Carlson DA, Quick M. Interaction of α-Lipoic Acid with the Human Na+/Multivitamin Transporter (hSMVT). J Biol Chem. 2015 Jun 26;290(26):16372-82. PMC free article
  2. Kobayashi M, Mizutani A, Okamoto T, Muranaka Y, Nishi K, Nishii R, Shikano N, Nakanishi T, Tamai I, Kleinerman ES, Kawai K. Assessment of drug transporters involved in the urinary secretion of [99mTc]dimercaptosuccinic acid. Nucl Med Biol. 2021 Mar-Apr;94-95:92-97. PubMed
  3. Peter G, Borbe HO. Absorption of [7,8-14C]rac-a-lipoic acid from in situ ligated segments of the gastrointestinal tract of the rat. Arzneimittelforschung. 1995 Mar;45(3):293-9. PubMed
  4. Uchida R, Okamoto H, Ikuta N, Terao K, Hirota T. Enantioselective Pharmacokinetics of α-Lipoic Acid in Rats. Int J Mol Sci. 2015 Sep 21;16(9):22781-94. PMC free article
  5. Aposhian HV, Aposhian MM. meso-2,3-Dimercaptosuccinic acid: chemical, pharmacological and toxicological properties of an orally effective metal chelating agent. Annu Rev Pharmacol Toxicol. 1990;30:279-306. PubMed
  6. Asiedu P, Moulton T, Blum CB, Roldan E, Lolacono NJ, Graziano JH. Metabolism of meso-2,3-dimercaptosuccinic acid in lead-poisoned children and normal adults. Environ Health Perspect. 1995 Jul-Aug;103(7-8):734-9. PMC free article
  7. van Eijkeren JC, Olie JD, Bradberry SM, Vale JA, de Vries I, Meulenbelt J, Hunault CC. Modelling dimercaptosuccinic acid (DMSA) plasma kinetics in humans. Clin Toxicol (Phila). 2016 Nov;54(9):833-839.
  8. Kitamura Y, Kusuhara H, Sugiyama Y. Functional characterization of multidrug resistance-associated protein 3 (mrp3/abcc3) in the basolateral efflux of glucuronide conjugates in the mouse small intestine. J Pharmacol Exp Ther. 2010 Feb;332(2):659-66. PubMed
  9. Kostial K, Blanusa M, Piasek M, Jones MM, Singh PK. Prolonged oral treatment with two monoesters of meso-2,3-dimercaptosuccinic acid for depleting inorganic mercury retention in suckling rats. Pharmacol Toxicol. 1995 Sep;77(3):216-8. PubMed
  10. Schepkin V, Kawabata T, Packer L. NMR study of lipoic acid binding to bovine serum albumin. Biochem Mol Biol Int. 1994 Aug;33(5):879-86 PubMed
  11. Naqvi S, Kumar P, Flora SJS. Comparative efficacy of Nano and Bulk Monoisoamyl DMSA against arsenic-induced neurotoxicity in rats. Biomed Pharmacother. 2020 Dec;132:110871 free article
  12. Chng HT, New LS, Neo AH, Goh CW, Browne ER, Chan EC. Distribution study of orally administered lipoic acid in rat brain tissues. Brain Res. 2009 Jan 28;1251:80-6. doi: 10.1016/j.brainres.2008.11.025. PubMed
  13. Schreibelt G, Musters RJ, Reijerkerk A, de Groot LR, van der Pol SM, Hendrikx EM, Döpp ED, Dijkstra CD, Drukarch B, de Vries HE. Lipoic acid affects cellular migration into the central nervous system and stabilizes blood-brain barrier integrity. J Immunol. 2006 Aug 15;177(4):2630-7. PubMed
  14. de Lange MJ, Piers DA, Kosterink JG, van Luijk WH, Meijer S, de Zeeuw D, van der Hem GK. Renal handling of technetium-99m DMSA: evidence for glomerular filtration and peritubular uptake. J Nucl Med. 1989 Jul;30(7):1219-23. PMC free article
  15. Siu B, Saha J, Smoyer WE, Sullivan KA, Brosius FC 3rd. Reduction in podocyte density as a pathologic feature in early diabetic nephropathy in rodents: prevention by lipoic acid treatment. BMC Nephrol. 2006 Mar 15;7:6 PMC free article
  16. Weyer K, Nielsen R, Petersen SV, Christensen EI, Rehling M, Birn H. Renal uptake of 99mTc-dimercaptosuccinic acid is dependent on normal proximal tubule receptor-mediated endocytosis. J Nucl Med. 2013 Jan;54(1):159-65. free article
  17. Zheng W, Maiorino RM, Brendel K, Aposhian HV. Determination and metabolism of dithiol chelating agents. VII. Biliary excretion of dithiols and their interactions with cadmium and metallothionein. Fundam Appl Toxicol. 1990 Apr;14(3):598-607. PubMed
  18. Thévenod F, Lee WK. Live and Let Die: Roles of Autophagy in Cadmium Nephrotoxicity. Toxics. 2015 Apr 13;3(2):130-151. PMC free article
  19. Chen XC, Li ZH, Yang C, Tang JX, Lan HY, Liu HF. Lysosome Depletion-Triggered Autophagy Impairment in Progressive Kidney Injury. Kidney Dis (Basel). 2021 Jul;7(4):254-267. PMC free article

Leave a Reply