One Enzyme, Three Jobs: Everything 5-Alpha-Reductase Makes

Finasteride and dutasteride are almost always described as "DHT blockers." That is accurate, but it is only one third of what the enzyme they block actually does.

The short answer: 5-alpha-reductase is famous as "the DHT enzyme," the target of finasteride and dutasteride. But that is only one of its jobs. The enzyme performs the same simple chemical step, adding two hydrogen atoms to a steroid, on at least three different families of molecules. Down the androgen lane it makes DHT, the hormone tied to scalp hair loss. Down the neurosteroid lane it helps build allopregnanolone, one of the brain's most powerful calming compounds. Down the glucocorticoid lane it helps clear cortisol, acting as part of a stress-hormone off-switch. That is why blocking the enzyme is discussed alongside mood and cortisol handling, not just hair. The androgen effect on hair is well established. The neurosteroid and cortisol effects are real biochemistry, but how much they matter clinically is still debated.

What does 5-alpha-reductase actually do?

5-alpha-reductase does one small chemical job, over and over, to different molecules. It takes a steroid with a particular double bond and reduces it, adding two hydrogen atoms across that bond (Russell and Wilson 1994). Chemically that is a minor edit. Biologically it can change everything about what the molecule does.

Humans carry more than one version of the enzyme. There are two classic isoforms, type 1 and type 2, encoded by separate genes and switched on in different tissues, plus a later-described type 3 (Azzouni et al 2012). The two classic isoforms split by tissue: type 1 predominates in scalp skin and sebaceous glands, while type 2 predominates in the hair follicle itself and in genital and prostatic tissue. Type 2 became the marquee drug target because of that prostatic and follicular role, but type 1 is not a bystander in the skin. And because all of these enzymes act on the shared steroid backbone, they do not touch androgens alone. The same reduction step lands on progesterone-family and cortisol-family steroids too.

The cleanest way to think about it: 5-alpha-reductase is not a hormone-specific switch. It is a general-purpose chemical tool that happens to sit at a fork in three different steroid highways.

Lane one: how does it make DHT and drive hair loss?

This is the pathway everyone knows. 5-alpha-reductase converts testosterone into dihydrotestosterone (DHT), and DHT binds the androgen receptor much more tightly than testosterone does (Azzouni et al 2012). Same receptor, stronger signal.

In scalp follicles that are genetically sensitive, that stronger androgen signal is associated with the follicle miniaturization of male pattern hair loss. In the beard and body, the same signal drives hair growth. It is the same molecule doing opposite-looking things in different follicles, which is one of the stranger facts in dermatology.

DHT is not the end of the road, either. It can be reduced and reshaped further into other molecules, including 3-alpha-androstanediol (3-alpha-diol) and 3-beta-androstanediol (3-beta-diol). These downstream products are interesting because they leave the androgen story entirely: 3-alpha-diol has been reported to interact with GABA-A receptors, and 3-beta-diol with estrogen receptor beta rather than the androgen receptor (Azzouni et al 2012). We flag these downstream steps as mechanistically described but not something with a settled role in hair-loss treatment.

The androgen-to-hair link is the strongest claim on this whole map. It is why 5-alpha-reductase inhibitors were developed for hair loss in the first place: finasteride 1 mg is FDA-approved for male androgenetic alopecia, and dutasteride is used off-label. This article does not make any efficacy or safety claim about either drug; it is about the biology they act on.

Lane two: how does the same enzyme help make a calming neurosteroid?

Here the story leaves hormones-as-hormones and enters the brain. Take progesterone, the reproductive steroid. 5-alpha-reductase reduces it to 5-alpha-dihydroprogesterone (5-alpha-DHP). A second enzyme, 3-alpha-hydroxysteroid dehydrogenase, then converts that into allopregnanolone (Azzouni et al 2012).

Allopregnanolone is not a reproductive hormone in any meaningful sense. It is a neurosteroid, and it is one of the most potent known positive modulators of the GABA-A receptor, the brain's main "slow down" system. In a classic study, Majewska and colleagues showed that steroid metabolites of this kind act on the GABA-A receptor in a barbiturate-like way, enhancing inhibitory signaling (Majewska et al 1986). In plain terms: this branch of the pathway helps produce a molecule that calms neural activity. A closely related steroid, deoxycorticosterone, is reduced along a parallel route to THDOC, another GABA-A-active neurosteroid.

The key point is that 5-alpha-reductase sits at the front of this calming pathway. That raises an obvious question about drugs that block the enzyme, and it is why 5-alpha-reductase inhibitors get discussed in the context of mood and neurosteroids at all. We want to be careful here: the biochemistry (enzyme, substrate, product, receptor) is well characterized, but the size and clinical importance of any mood effect from inhibiting this pathway in people is debated and not something we present as established fact.

Lane three: how does it act as a cortisol off-switch?

The third lane is the least famous and, biochemically, one of the neatest. Cortisol, the main human stress hormone, does not just float around until it decides to leave. The body actively breaks it down through several routes, and 5-alpha-reductase is one of the enzymes involved.

It helps to be precise about how large a route this is. The main off-switch for cortisol is 11-beta-hydroxysteroid dehydrogenase type 2, which converts cortisol to inactive cortisone. On top of that, the steroid A-ring is reduced by two competing enzymes: 5-beta-reductase, which handles the larger share, and 5-alpha-reductase. So 5-alpha-reduction is a real but minor tributary of cortisol clearance, not the master switch. 5-alpha-reductase (mainly the type 1 isoform) reduces cortisol to 5-alpha-dihydrocortisol, which is then processed onward to 5-alpha-tetrahydrocortisol (5-alpha-THF), an inactive metabolite headed for excretion (Hazlehurst et al 2016). Functionally, this reduction is one contributor to how the body clears cortisol.

That has a testable consequence: block the enzyme, and you should slow this particular cortisol-clearing route. Human data line up with that prediction. In men, dual 5-alpha-reductase inhibition altered cortisol handling and was associated with metabolic changes in the liver (Hazlehurst et al 2016). And in patients taking dutasteride, the ratio of 5-alpha to 5-beta steroid metabolites in the urine dropped, a direct fingerprint of the 5-alpha route being turned down while the parallel 5-beta route kept running (Maeda et al 2018). The enzyme really is doing measurable cortisol work in living people.

What we do not claim is that this translates into any specific symptom or clinical outcome for a person on these drugs. "The pathway is measurably affected" and "this causes outcome X in patients" are different statements, and only the first is well supported here.

So what does this mean for finasteride and dutasteride?

It reframes what a 5-alpha-reductase inhibitor is. These drugs are usually described as "DHT blockers," which is accurate but incomplete. Because the enzyme is a general-purpose steroid tool sitting at three forks, inhibiting it does not surgically remove DHT and leave everything else untouched. It can turn down a step that feeds more than one lane.

But the two approved drugs are not interchangeable here, and this is the part most summaries skip. Finasteride is largely selective for type 2. Dutasteride inhibits both type 1 and type 2. Because the three lanes rely on different mixes of the two isoforms, how much any given lane actually moves depends on which drug is used and which isoform runs that lane. The cortisol lane is a clear example: it is carried mainly by type 1, so it is the dual inhibitor dutasteride, not type-2-selective finasteride, that would be expected to turn it down most. That matches the evidence. The human cortisol-handling data on this map come from dual 5-alpha-reductase inhibition (Hazlehurst et al 2016) and from dutasteride specifically (Maeda et al 2018), not from finasteride. So "block the enzyme and every lane drops equally" is the wrong mental model. The honest version is: each lane moves by an amount set by its isoform mix and by the specific drug in question.

That is the honest mechanistic reason these medicines are discussed alongside neurosteroids and cortisol metabolism, and not only hair counts. Whenever you read about mood or metabolic effects tied to these drugs, this three-lane map, read with the isoform and drug caveats above, is the underlying biology being pointed at.

We are deliberately stopping at biology. How strong each downstream effect is, whether it matters for any individual, and how the benefits and risks balance are clinical questions that belong with a qualified prescriber and depend on evidence we are not adjudicating here. What this article establishes is narrower and, we think, genuinely useful: one enzyme, one simple chemical step, three very different destinations.

Frequently asked questions

Is 5-alpha-reductase only involved in hair loss?

No. Hair loss is downstream of just one of its jobs, the conversion of testosterone to DHT. The same enzyme also helps build calming neurosteroids from progesterone and helps clear cortisol, the stress hormone (Azzouni et al 2012, Hazlehurst et al 2016). It performs the same basic chemical step on several different steroid families.

What is allopregnanolone, and why does it come up with these drugs?

Allopregnanolone is a neurosteroid made partly by 5-alpha-reductase, and it is a strong positive modulator of the GABA-A receptor, the brain's main inhibitory system (Majewska et al 1986). Because the enzyme sits at the start of that pathway, blocking the enzyme is discussed in the context of neurosteroids. The clinical size of that effect in people is debated, not settled.

Does 5-alpha-reductase really affect cortisol?

Yes, as biochemistry. It reduces cortisol toward inactive metabolites (5-alpha-tetrahydrocortisol), part of how the body clears the hormone (Hazlehurst et al 2016). In people on dutasteride, the urinary 5-alpha to 5-beta metabolite ratio shifts, showing the pathway is genuinely turned down (Maeda et al 2018). Whether that produces any specific symptom is a separate, unproven question.

Are finasteride and dutasteride approved for hair loss?

Finasteride 1 mg is FDA-approved for male androgenetic alopecia. Dutasteride is used off-label for hair loss in the United States. This article is about the enzyme they target, not a recommendation or a claim about how well or how safely either drug works.

Why does DHT grow beard hair but shrink scalp hair?

The same DHT signal produces opposite responses in follicles with different genetic programming. Scalp follicles that are androgen-sensitive tend to miniaturize; beard and body follicles tend to be stimulated (Azzouni et al 2012). The hormone is identical; the follicle's response is what differs.

If the enzyme does three jobs, does blocking it hit all three?

Not equally, and the drug matters. Inhibiting the enzyme can touch more than the androgen lane, but each lane runs on a different mix of the type 1 and type 2 isoforms, and the two approved drugs differ: finasteride is largely type-2-selective, while dutasteride inhibits both. The cortisol lane is carried mainly by type 1, so its human data come from dual inhibition and dutasteride rather than finasteride.

References

1. Russell DW, Wilson JD. Steroid 5-alpha-reductase: two genes/two enzymes. Annu Rev Biochem. 1994;63:25-61. DOI: 10.1146/annurev.bi.63.070194.000325
2. Azzouni F, Godoy A, Li Y, Mohler J. The 5-alpha-reductase isozyme family: a review of basic biology and their role in human diseases. Adv Urol. 2012;2012:530121. DOI: 10.1155/2012/530121
3. Majewska MD, Harrison NL, Schwartz RD, Barker JL, Paul SM. Steroid hormone metabolites are barbiturate-like modulators of the GABA receptor. Science. 1986;232(4753):1004-1007. DOI: 10.1126/science.2422758
4. Hazlehurst JM, Oprescu AI, Nikolaou N, et al. Dual-5-alpha-reductase inhibition promotes hepatic lipid accumulation in man. J Clin Endocrinol Metab. 2016;101(1):103-113. DOI: 10.1210/jc.2015-2928
5. Maeda T, et al. Change of the 5-alpha/5-beta ratio of urinary steroid metabolites in benign prostatic hyperplasia patients treated with dutasteride. Clin Biochem. 2018;55:36-41. DOI: 10.1016/j.clinbiochem.2018.03.021

---

This article is educational and is not medical advice. It describes the biochemistry of an enzyme and the pathways it acts on, not the diagnosis or treatment of any condition, and it makes no claim about the efficacy or safety of any medication. Finasteride 1 mg is FDA-approved for male androgenetic alopecia; dutasteride is used off-label for hair loss in the United States. Talk to a qualified clinician about your own care.