What Are Phosphodiesterase Inhibitors and How Do They Work?

Phosphodiesterase inhibitors are a broad class of drugs that work by blocking enzymes called phosphodiesterases, which normally break down signaling molecules inside your cells. By preventing that breakdown, these drugs amplify the chemical signals that tell muscles to relax, blood vessels to widen, or immune cells to calm down. The result is a remarkably diverse medicine cabinet: the same basic mechanism underlies treatments for erectile dysfunction, chronic lung disease, heart failure, skin conditions, and pulmonary hypertension, among others. What makes the class so interesting, and so tricky to get right, is that there are eleven different families of phosphodiesterase enzymes, each with different jobs in different tissues, and targeting one family while leaving the others alone is the central challenge in designing these drugs.

How Phosphodiesterases Work and Why Blocking Them Matters

Your cells rely on internal messenger molecules, primarily cyclic AMP (cAMP) and cyclic GMP (cGMP), to relay signals from the outside world into action. When a hormone binds to a receptor on a cell’s surface, one of these messengers spikes inside the cell, triggering a cascade of effects: a heart muscle cell contracts more forcefully, a smooth muscle cell in a blood vessel relaxes, or an immune cell dials back inflammation. Phosphodiesterases are the enzymes that break down cAMP and cGMP, converting them into inactive forms and effectively turning the signal off.¹ Phosphodiesterase inhibitors jam that off-switch. The messenger molecules stick around longer, and the signal they carry gets louder and lasts longer.

This is a deceptively simple idea with complicated consequences. The same messenger molecule, cAMP, does different things in different cell types. Boosting cAMP in heart muscle makes it pump harder. Boosting cAMP in airway smooth muscle makes the airways relax and open up. Boosting cAMP in immune cells can suppress inflammation. A drug that raises cAMP everywhere in the body would produce all of these effects simultaneously, which is why the field has spent decades trying to develop drugs that are selective for specific phosphodiesterase families in specific tissues.

Eleven Families, One Shared Target

Mammals have 21 genes encoding phosphodiesterases, organized into 11 families (PDE1 through PDE11).² Some families primarily break down cAMP, others prefer cGMP, and a few break down both. They also differ in which tissues express them most heavily and where exactly within a cell they sit.³ PDE3, for instance, is abundant in heart muscle and blood vessel walls. PDE4 is the dominant family in immune cells and airway tissue. PDE5 is concentrated in the smooth muscle of the penis and the lung’s blood vessels. PDE10 is found almost exclusively in a brain region involved in movement and reward. These tissue-expression patterns are what make selective inhibitors useful: you can, in theory, design a drug that hits PDE5 hard without disturbing PDE3 in the heart.

In practice, selectivity is never perfect. Many drugs have “off-target” activity against related families, and those off-target effects explain a good chunk of the side effects patients experience. The story of phosphodiesterase inhibitors is, in many ways, the story of making selectivity sharper over time.

The Oldest Members of the Class

Long before anyone understood the enzyme families, people were using phosphodiesterase inhibitors without knowing it. Caffeine and theophylline are methylxanthines, a group of naturally occurring compounds that non-selectively inhibit multiple phosphodiesterase families. Theophylline was a mainstay of asthma treatment for decades precisely because it relaxes airway smooth muscle by raising intracellular cAMP. The trouble with non-selective inhibitors is side effects: theophylline can cause nausea, tremors, rapid heartbeat, and seizures at higher doses, because it is boosting cAMP and cGMP in tissues throughout the body rather than just in the airways. The push toward selective inhibitors was driven largely by the desire to keep the therapeutic effects and shed the side-effect burden of drugs like theophylline.

Erectile Dysfunction and PDE5 Inhibitors

The most publicly famous phosphodiesterase inhibitors are sildenafil (Viagra), tadalafil (Cialis), and vardenafil (Levitra), all of which target PDE5. Normal erections depend on nitric oxide being released in the penile tissue, which triggers a rise in cGMP, which in turn relaxes smooth muscle and allows blood to flow in.⁴ PDE5 is the enzyme that breaks down that cGMP, ending the erection. By blocking PDE5, these drugs keep cGMP levels elevated for longer after sexual stimulation, making it easier to achieve and maintain an erection.⁵

An important nuance: PDE5 inhibitors do not cause erections on their own. They amplify the body’s existing response to sexual arousal. Without the initial release of nitric oxide, there is nothing to amplify. This is why the drugs require sexual stimulation to work and why they are ineffective for certain types of erectile dysfunction that involve nerve damage severe enough to eliminate nitric oxide signaling entirely.

Sildenafil, famously, was originally being studied as a heart drug. Researchers at Pfizer were investigating it for angina in the early 1990s. The cardiac effects turned out to be modest, but trial participants reported a notable side effect: improved erections. The rest is pharmaceutical history.

Pulmonary Arterial Hypertension

PDE5 is not only found in penile tissue. It is also heavily expressed in the smooth muscle of pulmonary arteries, and this opened an entirely separate use for the same drugs. Pulmonary arterial hypertension (PAH) is a condition where the blood vessels in the lungs become abnormally constricted, forcing the right side of the heart to work dangerously hard. By relaxing those vessels, PDE5 inhibitors lower pulmonary artery pressure and improve blood flow through the lungs.⁶

Sildenafil was approved for PAH under the brand name Revatio, and tadalafil followed under the brand name Adcirca. In a large trial, patients receiving tadalafil at its approved dose walked about 33 meters farther in six minutes than those on placebo, a standard measure of exercise capacity in PAH research, and also showed improvement in quality-of-life measures.⁷ That may sound modest, but for patients who become breathless walking across a room, it represents a meaningful change. PDE5 inhibitors are now one of three major drug classes used for PAH alongside prostanoids and endothelin receptor antagonists.⁸

Heart Failure and the PDE3 Paradox

PDE3 inhibitors such as milrinone and amrinone work on the heart by a parallel logic: they raise cAMP levels in cardiac muscle cells, which increases the concentration of intracellular calcium and makes the heart contract more forcefully.⁹ They also relax blood vessels, reducing the workload on the heart. In the short term, this combination is genuinely useful: milrinone is still given intravenously in hospitals for acute heart failure episodes when the heart needs a temporary boost.

The paradox is that long-term oral use of PDE3 inhibitors has generally increased mortality in heart failure patients rather than decreasing it.¹⁰ The exact reasons remain unclear, but the most likely explanation involves the same mechanism that makes the drugs work: sustained cAMP elevation in the heart can promote dangerous rhythm disturbances. Multiple large trials in the 1980s and 1990s confirmed the increased risk, and long-term PDE3 inhibitor use for heart failure has essentially been abandoned as a strategy. It remains one of cardiology’s cautionary tales about the gap between making the heart feel stronger and actually keeping patients alive longer.

Airway Disease and PDE4 Inhibitors

PDE4 is the dominant phosphodiesterase in immune cells like neutrophils, eosinophils, and macrophages, as well as in airway smooth muscle. Blocking PDE4 has two effects that matter for lung disease: it relaxes the airways and it dials down the inflammatory response that drives conditions like chronic obstructive pulmonary disease (COPD). Roflumilast is the best-known selective PDE4 inhibitor approved for COPD.¹¹ It works partly by directly blocking neutrophil migration toward the inflamed tissue, reducing the immune cell traffic that causes ongoing lung damage.¹²

The main clinical limitation with PDE4 inhibitors taken by mouth is gastrointestinal side effects. PDE4 is also present in the gut, and inhibiting it there tends to cause nausea, diarrhea, and weight loss. These side effects are significant enough that roflumilast is typically reserved for patients with severe COPD and frequent flare-ups who are already on standard inhaled medications. The side-effect profile is far better when PDE4 inhibitors are applied directly to the skin, which is why the same mechanism has been more comfortably exploited in dermatology.

Skin Conditions and Topical PDE4 Inhibitors

Atopic dermatitis, psoriasis, and seborrheic dermatitis all involve inflammatory pathways that PDE4 inhibition can quiet. Crisaborole is a topical PDE4 inhibitor approved for atopic dermatitis in both children and adults. In two large trials, patients using crisaborole ointment were significantly more likely to achieve clear or almost-clear skin compared to those using a vehicle ointment, and they saw improvement in itch faster as well.¹³ Other PDE4 inhibitors, including apremilast (taken orally for psoriasis and psoriatic arthritis) and topical roflumilast for plaque psoriasis and seborrheic dermatitis, have also received FDA approval.¹⁴

These drugs fill a niche that has long been underserved. The standard topical treatment for eczema and psoriasis is corticosteroids, which work well but carry risks with prolonged use including skin thinning. PDE4 inhibitors offer a steroid-free alternative for maintenance therapy, particularly in sensitive areas like the face and skin folds where steroid side effects are most concerning. They are not as potent as strong topical steroids for acute flares, but their safety profile makes them attractive for long-term management.

The Dangerous Interaction with Nitrates

The single most important safety issue with PDE5 inhibitors is their interaction with nitrate medications used for chest pain, such as nitroglycerin and isosorbide mononitrate. Both nitrates and PDE5 inhibitors act on the same signaling pathway: nitrates flood the system with nitric oxide, which raises cGMP, and PDE5 inhibitors prevent that cGMP from being broken down. Together, they produce a massive, sustained drop in blood pressure that can be life-threatening.¹⁵ PDE5 inhibitors are contraindicated in anyone taking organic nitrates, and this interaction creates a genuine clinical problem: a man using sildenafil for erectile dysfunction who then develops chest pain cannot safely receive nitroglycerin in the emergency room.¹⁶ Emergency physicians have to wait for the PDE5 inhibitor to clear the body before administering nitrates, which can delay treatment for a heart attack.

Visual Side Effects and PDE6 Cross-Reactivity

All PDE5 inhibitors weakly inhibit PDE6, an enzyme found exclusively in the rod and cone cells of the retina that plays a central role in converting light into electrical signals.¹⁷ This cross-reactivity explains one of the more unusual side effects of sildenafil: a temporary blue-green tint to vision, increased sensitivity to light, and blurred vision. These visual symptoms are typically mild, transient, and correlate with the drug’s peak plasma concentration.¹⁸ At standard doses, the effect is harmless and clears within hours. At very high doses or in people with pre-existing retinal conditions, it may be more pronounced. Tadalafil, which is chemically distinct from sildenafil and vardenafil, tends to have less PDE6 cross-reactivity and correspondingly fewer visual complaints.

This is a clean example of why selectivity matters. PDE5 and PDE6 are structurally similar enough that a molecule designed to fit one will partially fit the other. Getting the chemical structure exactly right to distinguish between them is a core problem in drug design. Crystal structures of the PDE5 catalytic domain bound to sildenafil, tadalafil, and vardenafil have given researchers a detailed three-dimensional map for designing the next generation of more selective inhibitors.¹⁹

Brain Research and PDE Inhibitors

Some of the most intriguing research on phosphodiesterase inhibitors is happening in neurology, though very little of it has reached the clinic yet. PDE10A is concentrated in the striatum, the brain region that degenerates in Huntington’s disease. In mouse models of Huntington’s, PDE10A inhibitors corrected abnormal firing patterns in brain circuits, reduced the loss of brain cells, and improved motor function.²⁰ In one study, early chronic treatment delayed the progression of neurological deficits and partially reversed the abnormal gene-expression patterns typical of the disease.²¹ These are animal results, and the history of Huntington’s drug development is littered with compounds that worked in mice but failed in humans, so caution is warranted.

Alzheimer’s disease research has explored inhibitors of PDE2, PDE4, PDE5, and PDE9. In animal models of Alzheimer’s, inhibitors of PDE2, PDE4, and PDE5 have all improved memory performance.²² PDE2 inhibitors are a particularly active area: new compounds have recently shown the ability to reverse memory deficits in mouse Alzheimer’s models, working through a signaling chain that ultimately boosts a nerve-growth factor called BDNF.²³ However, clinical results have been disappointing so far. PDE9 inhibitors that looked promising in preclinical work failed to show cognitive benefits in patients with mild cognitive impairment or Alzheimer’s.²⁴ The gap between animal data and human outcomes remains the fundamental problem.

Kidney Disease

A somewhat unexpected application for phosphodiesterase inhibitors is in protecting the kidneys of people with diabetes. Pentoxifylline, an older non-selective methylxanthine derivative, has shown significant effects in reducing protein leakage into the urine (a marker of kidney damage) and slowing the loss of kidney function in diabetic kidney disease.²⁵ Meanwhile, a selective PDE5 inhibitor called PF-00489791 reduced urine albumin levels by about 16% compared to placebo over twelve weeks in patients with diabetic nephropathy, suggesting that the cGMP pathway may also matter for kidney health.²⁶ Neither approach has yet produced the definitive outcome data needed for broad adoption, but the idea of repurposing existing drugs for kidney protection is attractive given how few good options exist for slowing diabetic kidney disease beyond standard blood-pressure and blood-sugar control.

Cancer Research and PDE5

Emerging laboratory work has explored whether PDE5 inhibitors could make cancer cells more vulnerable to chemotherapy. In prostate cancer cell lines, sildenafil and vardenafil (but not tadalafil) enhanced the cell-killing effects of the chemotherapy drug doxorubicin by interfering with the cancer cell’s ability to repair DNA damage.²⁷ The interesting wrinkle is that this effect appeared to be independent of PDE5 itself, meaning sildenafil was hitting some other target in the cancer cells. Tadalafil, which has a very different chemical structure despite targeting the same enzyme, did not produce the same effect, reinforcing the idea that the anticancer activity was structural rather than mechanism-based. This remains firmly in the laboratory phase and is a long way from clinical use, but it illustrates how drugs designed for one purpose can reveal unexpected biology.

Tolerance and Cellular Adaptation

One question that comes up with any long-term medication is whether the body adapts and the drug loses effectiveness. There is evidence that cells can fight back against sustained PDE inhibition. In one laboratory study, long-term exposure of endothelial cells to the PDE4 inhibitor rolipram actually caused the cells to increase their production of PDE4 enzymes, apparently as a compensatory mechanism to get cAMP levels back to normal.²⁸ In other words, the cells upregulated the very enzyme the drug was trying to block. Whether this happens to a clinically meaningful degree in patients taking PDE4 or PDE5 inhibitors long-term is not fully established, but the basic biology suggests the body does try to resist the drug’s effects over time. For PDE5 inhibitors used in erectile dysfunction, dose escalation over the years is common in clinical practice, though age-related progression of the underlying vascular disease makes it hard to separate true pharmacological tolerance from worsening of the condition being treated.

Natural Phosphodiesterase Inhibitors

Beyond caffeine and theophylline, researchers have screened hundreds of plant-derived compounds for phosphodiesterase-inhibiting activity. The most effective natural PDE5 inhibitor identified to date is icariin, a compound found in the plant Epimedium brevicornum, commonly known as horny goat weed, a name that makes more sense once you know it contains a PDE5 inhibitor.²⁹ Dozens of flavonoids and alkaloids from various plants show some degree of PDE5 inhibition in laboratory assays, with prenylated isoflavones and certain biflavones being particularly active. None approach the potency of pharmaceutical PDE5 inhibitors, however, and the concentrations needed for meaningful effects in a living person would far exceed what you could reasonably consume from supplements or traditional preparations. The research is more useful as a starting point for drug design than as a practical alternative to prescription medication.

How Drug Designers Improve Selectivity

The structural similarities among phosphodiesterase families are what make selectivity difficult but not impossible. Structural biology studies have shown that all PDE inhibitors, despite their very different chemical structures, share a common binding mode: a flat ring system that slots into a hydrophobic pocket in the enzyme’s active site and forms a hydrogen bond with a specific glutamine residue that is conserved across all PDE families.³⁰ Selectivity comes from the decorations around that core ring. Different PDE families have slightly different shapes in the regions surrounding the active site, and adding chemical groups that exploit those differences is how medicinal chemists steer a drug toward one family and away from others. The crystal structures of PDE5 bound to sildenafil, tadalafil, and vardenafil have been especially valuable for this work, revealing exactly which parts of each molecule contact which parts of the enzyme.³¹ The long-term goal is inhibitors so selective they produce therapeutic effects in one tissue without meaningful off-target activity anywhere else, though that ideal has not yet been fully achieved for any PDE family.