Ambroxol and the Blonde Rat: the Rab38 Connection

By Brian · 19 min read · Updated September 14, 2026

This is the deepest and most personal of our open questions, the one we have chased the hardest. The gene that makes a roof rat blonde, Rab38, also runs the exact cellular machinery that the drug ambroxol acts on. Most of the pieces are already in the published literature. The single missing step is whether they connect in a living rat, and our blonde colony is unusually well placed to find out. Here is the whole story, from the coat down to the lysosome.

A note before we start: this is a research hypothesis, not medical advice. It is not a reason to dose a blonde rat differently. Ambroxol stays an adjunct to veterinary antibiotics, at your vet’s dose, for any rat.

What Rab38 actually builds

Rab38 is a small trafficking switch, and its job is to deliver cargo to a special family of compartments called lysosome-related organelles. These are the cell’s customised storage-and-release bags. The same gene, the rat Ruby locus, is the one mutated in chocolate mice and fawn-hooded rats, and it is what our blonde roof rats are missing.^[1][2] What is striking is how few jobs Rab38 has, and how visible they are:

• Melanosomes, the pigment bags. Lose Rab38 and pigment is diluted, which is literally why a blonde rat is blonde.^[1]
• Lamellar bodies, the lung’s surfactant bags. Rab38 is active in only two cell types in the whole body, pigment cells and lung alveolar type II cells, where it keeps the lamellar bodies the right size and working.^[3]
• Platelet dense granules, which help blood clot. This is the root of the blonde bleeding caution we put on the supplement pages, and it is its own open question.^[4]

A close cousin, Rab32, backs Rab38 up and softens the blow of losing it, which is part of why a Rab38-null rat is dilute rather than dead.^[5] The key idea to carry forward is this: a blonde rat’s deficiency is, at its root, a lysosome-related-organelle problem. The coat is just the part you can see.

That backup is not the same everywhere in the body, and the difference turns out to matter for rats specifically. Mouse and human platelets express Rab32 alongside Rab38, so a mouse missing only Rab38 keeps normal dense granules. Rat platelets do not express Rab32 at all. When Aguilar and colleagues knocked out both genes in mice, the double-null animals developed strongly diluted coat and eye pigment, some enlarged lung multilamellar bodies, fewer and structurally abnormal dense granules with no serotonin content, impaired thrombus growth and a prolonged bleeding time.^[6] That is the picture a rat reaches with a Rab38 defect alone, because it has no Rab32 to fall back on in the platelet. It is the cleanest published explanation for why the Ruby-locus rat and the chocolate mouse do not behave the same way.

The dense-granule job is worth one more sentence, because it shows what kind of machine Rab38 is. Working in a human megakaryocyte cell line, Ambrosio and colleagues traced dense granules to the late endocytic route and showed that the sorting signals read by adaptor protein-3 are needed to deliver cargo there, with Rab32 and Rab38 required for the step where cargo-carrying vesicles fuse with the maturing granule.^[7] Rab38 is not building the bag from scratch. It is the switch that lets the last delivery dock.

There is also a mechanical reason a Rab38 mutation can be total rather than partial. In the chocolate mouse, described by Osanai and colleagues in 2008, the mutant Rab38 protein is still made and still binds GTP, but it fails to pick up the lipid tail (prenyl modification) that a Rab needs in order to sit on a membrane, so it stays loose in the cytoplasm and never reaches the organelle it is supposed to steer.^[8] A switch that is manufactured correctly but never installed is functionally the same as no switch at all.

The Ruby-locus source note links to Oiso and colleagues’ identification of Rab38 in Fawn-hooded and Tester-Moriyama laboratory rats.^[1]

The lung, where it gets interesting

Of all those jobs, the lung is the one that matters most for a sick rat. Surfactant is the soapy film that keeps the tiny air sacs from collapsing on every breath, and it is stored and released from lamellar bodies, the same bags Rab38 manages.

In Rab38-null rats this is not subtle. The lamellar bodies swell to about 77 percent larger than normal, surfactant handling is disrupted, and the air-sac architecture is abnormal. Put a working copy of Rab38 back, and it rescues.^[3][9] So a blonde, dilute, Rab38-deficient rat may carry a built-in surfactant disadvantage, in exactly the system that respiratory disease attacks and that ambroxol acts on.

The mouse version of the same defect shows what is actually going wrong, and it is not what most people would guess. Chocolate mice, which carry a Rab38 point mutation, have enlarged distal airspaces, mild alveolar destruction and a slight rise in lung compliance. Their alveolar type II cells are engorged with lamellar bodies that are both larger and more numerous than normal. The surfactant itself is not missing: phosphatidylcholine and surfactant protein B are increased in the lung tissue while being decreased in the air spaces where they are needed.^[8] So a Rab38 defect is better read as a delivery failure than a manufacturing failure. The factory keeps running and the loading dock jams, which is exactly the kind of problem a drug that pushes lysosome-related organelles to secrete might be able to unjam.

The studies separate two kinds of evidence. Zhang and colleagues restored Rab38 expression in cultured cells from Fawn-hooded hypertension rats and rescued the enlarged lamellar-body phenotype.^[3] Osanai and colleagues used Rab38 gene transfer in Long Evans Cinnamon rats, improving surfactant secretion patterns and reducing the enlarged organelles.^[9] Those were gene-restoration experiments, not tests of ambroxol treatment in blonde roof rats.

What ambroxol really does, beyond thinning mucus

Most people know ambroxol as a cough medicine. Its deeper pharmacology is the interesting part, and it lines up with Rab38’s world point for point:

• It changes surfactant-protein expression. In type II pneumocytes isolated from ambroxol-treated Sprague-Dawley rats, SP-C protein and mRNA increased.^[10]
• It is a lysosomal-enzyme chaperone. Ambroxol binds the enzyme glucocerebrosidase and helps fold and traffic it into the lysosome, which is why it is in human trials for Parkinson’s and Gaucher disease.^[11][12]
• It works at the whole-animal level. Oral ambroxol crosses into the brain and raises glucocerebrosidase activity in a living primate, so this is not just a cell-dish effect.^[13]
• It changes lysosomal and secretory pathways. In cultured mouse cortical neurons, ambroxol increased TFEB and several lysosomal enzymes, but macroautophagy flux was blocked while exocytosis increased.^[14]

Read that list again next to Rab38’s. Ambroxol’s published job is, almost line for line, to boost lysosomal biogenesis, chaperone lysosomal proteins, and drive lysosome-related organelles to secrete. Rab38’s job is to traffic cargo to those same organelles.

One study gets closer to the overlap than any of the others, and it is the reason this page exists at all. Fois and colleagues used X-ray spectroscopy to show that ambroxol physically accumulates inside lamellar bodies, the surfactant-storing secretory lysosomes of type II pneumocytes. Once inside, it acts as a weak base and raises the pH of the compartment, which releases calcium from those acidic stores and produces dose-dependent surfactant exocytosis. Blocking the vesicular proton pump with bafilomycin A1 reduced and slowed the calcium release, and the authors also recorded a change in how the surfactant layers assemble. They describe ambroxol as a lysosomal secretagogue.^[15] Read that against the surfactant source note: the drug concentrates inside the exact organelle that Rab38 exists to build, and then tells it to empty.

The surfactant effect is narrower than the phrase “it boosts surfactant” suggests, and the detail is worth having. Seifart and colleagues treated Sprague-Dawley rats with ambroxol and then looked separately at isolated type II pneumocytes, at whole lung tissue and at lavage fluid. In the type II cells, surfactant protein C rose in both protein and messenger RNA while surfactant proteins A, B and D were unaffected. In whole lung tissue, surfactant protein B rose and surfactant protein D fell, with the extra B traced by staining to Clara cells rather than type II cells. In the lavage fluid, both A and D fell.^[10] So ambroxol does not turn the surfactant system up as a whole. It moves particular proteins in particular cells, in different directions, which is the kind of result that is easy to lose when a mechanism is summarised in four words.

The anti-inflammatory half of the story is the best-supported half in a live animal, and it is worth being exact about what was measured. In a mouse model of acute lung injury, ambroxol given by injection at 30 or 90 mg/kg per day for seven days reduced lung haemorrhage, oedema, neutrophil infiltration and the total injury score, lowered tumour necrosis factor alpha, interleukin-6 and transforming growth factor beta-1 in lavage fluid, and cut the protein leak across the vessel wall.^[16] Separately, ambroxol scavenges hydroxyl radicals and hypochlorous acid directly in a test tube, and modestly reduces superoxide, although it does nothing to hydrogen peroxide and needs high concentrations to do any of it.^[17] Both strands are collected in the anti-inflammatory source note. Neither was done in a rat with a respiratory infection, and neither involved a Rab38 mutant.

The surfactant source note distinguishes protein expression from secretion: Seifart and colleagues measured different responses in different cell types, including increased SP-C in type II pneumocytes and enhanced SP-B staining in Clara cells.^[10] The lysosomal-chaperone source note brings together patient-derived skin fibroblasts and a separate healthy nonhuman-primate study: the former showed increased lysosomal mutant GCase and enzyme activity, while the latter measured increased brain GCase activity.^[11][13] These are different experimental systems and endpoints, not a demonstrated rescue of the blonde-rat phenotype.

The distinction between more lysosomal machinery and faster breakdown matters here. In the primary cortical-neuron study, higher lysosomal-enzyme and TFEB levels occurred alongside blocked macroautophagy flux and increased exocytosis. The authors interpreted this as cargo being diverted toward secretion.^[14] It should not be read as a general demonstration that ambroxol makes cells digest more material.

Where they meet: the compensation question

So here is the question this whole page is built around, stated as a clean experiment. It is the one we would most like to see someone run.

Why the blonde is a clean model

A drug whose published job is to prop up the lysosome-related-organelle system, meeting an animal whose defining trait is a lysosome-related-organelle gene knocked out, is a natural experiment. And our blonde is a good one to run it in: the Rab38 loss is PCR-confirmed, not inferred from coat alone, and an outside genetics lab has now independently characterized that same deletion and presented it (Kido and Kuramoto, Tokyo University of Agriculture, at the 2024 Annual Meeting of the Zoological Society of Japan; their peer-reviewed paper is in preparation).^[18] It is carried in a tame, breedable colony. That makes it a living test bed for ambroxol across trafficking and lysosomal-storage biology, not just the lung. It also sits inside our wider coat-defined allelic series (agouti, blonde, and black), which is itself a rare genetics resource.

The useful thing about that literature is that it has already written the measurement list. The Rab38 papers score lamellar-body size and number in alveolar type II cells, and they score how surfactant is split between lung tissue and air space, which is the split that goes wrong in a Rab38 mutant.^[8][3] The two gene-transfer studies then show what a successful rescue looks like on those same measures.^[9][3] An ambroxol arm would therefore not need a new assay. It would need the existing assay run on treated and untreated Rab38-null animals side by side, with wild-type controls, which is a large part of why we keep describing this as a fundable experiment rather than a wish.

The honest dose problem

Here is the catch, and it matters. Ambroxol does different things at different doses. At the modest oral dose a pet rat actually gets, the realistic effect is anti-inflammatory and mildly mucokinetic. The deep effects, robust surfactant induction and strong glucocerebrosidase chaperoning, were measured at much higher doses, and the pain-blocking sodium-channel effect needs higher doses still.^[19][10] So the compensation idea is a hypothesis about biology, tested at research doses, not a reason to push more ambroxol into your blonde rat. Both halves of the story, Rab38’s role in lamellar bodies and ambroxol’s lysosomal action, are in the peer-reviewed literature. The single untested step is whether the one meaningfully rescues the other in a living rat. That is a clean, fundable experiment, and ambroxol remains a comfort adjunct to veterinary antibiotics either way.^[20]

It helps to put real numbers on “much higher doses”, because the published rat work uses a narrow set of them. Two 2026 rat studies both dosed oral ambroxol by daily gavage at 10, 50 and 100 mg/kg, one for 21 days and one for 28 days.^[21][22] The bottom rung of that ladder, 10 mg/kg per day, is roughly the dose a pet rat is given. The anti-inflammatory mouse work used 30 and 90 mg/kg per day by injection.^[16] The rat study that found the surfactant-protein changes gave 75 mg/kg by injection twice a day, which is about 150 mg/kg per day by a route that skips the gut entirely.^[10] The direct antioxidant effects were measured in a dish at concentrations the paper itself calls high.^[17] And the pain-blocking effect has a specific mechanism behind it: ambroxol binds the same site on the sodium channel that local anaesthetics use, with its use-dependent block stronger on Nav1.8 than on the tetrodotoxin-sensitive channels, which is why it behaves like a mild local anaesthetic rather than a painkiller in the usual sense.^[23] That mechanism is collected in the local-anaesthetic source note.

Route changes the picture as much as dose does, which is the part of the problem most within a keeper’s reach. In rats given 20 mg/kg of ambroxol as a dry powder into the trachea, the ratio of drug exposure in the lung’s epithelial lining fluid to exposure in plasma ran between 1.05 and 2.25. The same dose given intravenously produced a ratio between 0.029 and 0.039.^[24] In other words, where the drug is put decides how much of it reaches the airway surface, by a wide margin. That is the published basis for the open question about nebulised versus oral dosing, and it is one reason the dose ceiling on this page is a statement about the oral route rather than about the molecule. The general pharmacology behind all of this is gathered in the ambroxol pharmacology review note.

The “helps antibiotics reach the lung” idea deserves the same treatment, because the review everyone cites for it is more careful than the shorthand. Deretic and Timmins report that ambroxol co-treatment raises lung and airway-surface levels of beta-lactams, glycopeptides, macrolides, nitrofurans and rifamycins. They then say the increases are in most cases only modest, insufficient to overcome high-level resistance, and unlikely to change clinical outcomes, and that co-treatment has limited efficacy against most pathogens. The two situations they single out as genuinely promising are rifampin-sensitive tuberculosis, non-tuberculous mycobacterial infection, and vancomycin-sensitive MRSA pneumonia.^[20] None of the classes on that list is doxycycline or enrofloxacin, which are the drugs a rat with a respiratory infection is usually given. Ambroxol earns its place here as a comfort and surfactant adjunct, and the case for it as a delivery booster for the specific antibiotics rats receive has not been made.

One last piece of honesty, since this page is enthusiastic about a drug. Ambroxol has a long over-the-counter record outside the United States, and in 2015 the European Medicines Agency reviewed ambroxol and bromhexine medicines for severe allergic reactions and severe skin reactions. It judged the risk small and kept the medicines available, with the product information updated to name those reactions.^[25] Small is not zero, and a review that finds a real signal is worth reading before treating any drug as harmless.

Common questions

Should I give my blonde rat more ambroxol than other rats?

No. This page is about a research idea, not a dosing instruction. At the doses a pet rat actually gets, ambroxol works mostly as an anti-inflammatory, and you should follow your vet’s dose, which is the same whatever colour your rat is. Nothing here is a reason to dose a blonde rat differently.

Is the Rab38 lung connection in blonde rats proven?

The two halves are published and solid: Rab38 runs the lamellar bodies that store and release lung surfactant, and ambroxol acts on those same lamellar bodies and on the wider lysosomal system. The one untested step is whether ambroxol actually compensates for Rab38 loss in a living rat. That is the open experiment.

Why would a blonde rat be a good model for this?

Because the blonde coat is a confirmed Rab38 null carried in a tame, breedable line. That makes it a living test bed for ambroxol’s effects on lysosome-related organelles, in the lung and beyond, which is hard to find anywhere else.

Does this mean blonde rats have worse lungs?

Not that we see in daily life. It is a hypothesis that a Rab38-deficient rat may carry a built-in surfactant disadvantage worth measuring, not a diagnosis. If anything, it is a reason the surfactant side of ambroxol might matter more for a blonde rat, not less.

Go deeper

Related: Ambroxol for rats (the practical page) · The ergothioneine deep dive (the other Rab38 story) · Coat colour and genetics · All open research questions.

This page describes a research hypothesis and the published science around it, not veterinary advice. We have nothing to sell. The colony exists to keep these rats and this knowledge going.

References

• [1] Oiso N et al. “The rat Ruby (R) locus is Rab38: identical mutations in Fawn-hooded and Tester-Moriyama rats derived from an ancestral Long Evans rat sub-strain.” Mammalian Genome 2004;15(4):307-314. PubMed.
• [2] Loftus SK et al. “Mutation of melanosome protein RAB38 in chocolate mice.” Proceedings of the National Academy of Sciences 2002;99(7):4471-4476. PubMed.
• [3] Zhang L et al. “Rab38 targets to lamellar bodies and normalizes their sizes in lung alveolar type II epithelial cells.” American Journal of Physiology. Lung Cellular and Molecular Physiology 2011;301:L461-L477. PubMed.
• [4] Ninkovic I et al. “The role of Rab38 in platelet dense granule defects.” Journal of Thrombosis and Haemostasis 2008;6(12):2143-2151. PubMed.
• [5] Wasmeier C et al. “Rab38 and Rab32 control post-Golgi trafficking of melanogenic enzymes.” Journal of Cell Biology 2006;175(2):271-281. PubMed.
• [6] Aguilar A, Weber J, Boscher J, et al. “Combined deficiency of RAB32 and RAB38 in the mouse mimics Hermansky-Pudlak syndrome and critically impairs thrombosis.” Blood Advances 2019;3(15):2368-2380. PubMed.
• [7] Ambrosio AL, Boyle JA, Di Pietro SM. “Mechanism of platelet dense granule biogenesis: study of cargo transport and function of Rab32 and Rab38 in a model system.” Blood 2012;120(19):4072-4081. PubMed.
• [8] Osanai K, Oikawa R, Higuchi J, Kobayashi M, et al. “A mutation in Rab38 small GTPase causes abnormal lung surfactant homeostasis and aberrant alveolar structure in mice.” American Journal of Pathology 2008;173:1265-1274. PubMed.
• [9] Osanai K et al. “Exogenous gene transfer of Rab38 small GTPase ameliorates aberrant lung surfactant homeostasis in Ruby rats.” Respiratory Research 2017;18:70. PubMed.
• [10] Seifart C et al. “Cell-specific modulation of surfactant proteins by ambroxol treatment.” Toxicology and Applied Pharmacology 2005;203:27-35. PubMed.
• [11] Bendikov-Bar I et al. “Ambroxol as a pharmacological chaperone for mutant glucocerebrosidase.” Blood Cells, Molecules & Diseases 2013;50:141-145. PubMed.
• [12] McNeill A, Magalhães J, Shen C, et al. “Ambroxol improves lysosomal biochemistry in glucocerebrosidase mutation-linked Parkinson disease cells.” Brain 2014;137(Pt 5):1481-1495. PubMed.
• [13] Migdalska-Richards A, Ko WKD, Li Q, et al. “Oral ambroxol increases brain glucocerebrosidase activity in a nonhuman primate.” Synapse 2017;71(7):e21967. PubMed.
• [14] Magalhães J et al. “Effects of ambroxol on the autophagy-lysosome pathway and mitochondria in primary cortical neurons.” Scientific Reports 2018;8:1385. PubMed.
• [15] Fois G, Hobi N, Felder E, Ziegler A, et al. “A new role for an old drug: Ambroxol triggers lysosomal exocytosis via pH-dependent Ca2+ release from acidic Ca2+ stores.” Cell Calcium 2015;58:628-637. PubMed.
• [16] Su X et al. “Inhibition of inflammatory responses by ambroxol, a mucolytic agent, in a murine model of acute lung injury induced by lipopolysaccharide.” Intensive Care Medicine 2004;30:133-140. PubMed.
• [17] Gillissen A et al. “Oxidant scavenger function of ambroxol in vitro: a comparison with N-acetylcysteine.” Research in Experimental Medicine 1997;196:389-398. PubMed.
• [18] Kido M, Tanaka K, Kimura G, Kuramoto T. “A fawn coat-colour mutant found in captive roof rats (Rattus rattus) carries a deletion in the Rab38 gene [conference presentation, in Japanese; original title 飼育クマネズミに見出された淡毛色変異体はRab38遺伝子に欠失を持つ].” 95th Annual Meeting of the Zoological Society of Japan, Nagasaki; presentation 1A0900 (Mammals/Genetics session), Tokyo University of Agriculture 2024. source.
• [19] Gaida W et al. “Ambroxol, a Nav1.8-preferring Na+ channel blocker, effectively suppresses pain symptoms in animal models of chronic, neuropathic and inflammatory pain.” Neuropharmacology 2005;49:1220-1227. PubMed.
• [20] Deretic V, Timmins GS. “Enhancement of lung levels of antibiotics by ambroxol and bromhexine.” Expert Opin Drug Metab Toxicol 2019;15(3):213-218. PubMed.
• [21] Frota GM, Santos WGB, Tenorio-Meireles J, et al. “Analgesic and Anti-Inflammatory Activity of Ambroxol in the Treatment of Endometriosis: An Experimental Study in Wistar Rats.” Pharmaceuticals (Basel) 2026;19(4):641. PubMed.
• [22] Mesquita MJTAM, Cruz ACSND, Teixeira JTAMMM, et al. “Therapeutic Potential of Ambroxol in Osteoarthritis: A Drug Repurposing Study.” Pharmaceuticals (Basel) 2026;19(5):677. PubMed.
• [23] Leffler A, Reckzeh J, Nau C. “Block of sensory neuronal Na+ channels by the secreolytic ambroxol is associated with an interaction with local anesthetic binding sites.” European Journal of Pharmacology 2010;630:19-28. PubMed.
• [24] Ren YC, Wang L, He HB, Tang X. “Pulmonary selectivity and local pharmacokinetics of ambroxol hydrochloride dry powder inhalation in rat.” J Pharm Sci 2009;98(5):1797-1803. PubMed.
• [25] European Medicines Agency (PRAC). “Ambroxol- and bromhexine-containing medicines: Article 31 referral (2015 PRAC review of severe allergic reactions).” link.