What is Dr. Zubair Khalid's research focus?

Dr. Zubair Khalid specializes in molecular virology, mRNA vaccine development, and computational biology, with a focus on avian pathogens like IBDV and Avian Reovirus.

Where is Dr. Zubair Khalid currently working?

Dr. Zubair Khalid is a Postdoctoral Research Associate at the University of Maryland (UMD), specifically within the Department of Animal and Avian Sciences.

What Does Persistence Mean in Biology and Medicine

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-09-20

Persistence is the ability of a biological entity (a cell, a population, a virus, or a drug molecule) to remain present in a system over time despite forces that would normally remove, kill, or clear it. In medicine, the term almost always signals a survival problem: something that should have been eradicated is still there.

That single idea runs through microbiology, pharmacology, and ecology, but it wears a different costume in each field. A persister bacterium in a biofilm, a latent herpesvirus in a sensory neuron, a drug bound to a receptor hours after plasma levels fall, and a small fish population hanging on in marginal habitat are all examples of persistence. The mechanisms differ enormously. The logic is the same: staying power under pressure.

This guide defines persistence in each domain, separates it from resistance and from related terms like latency and tolerance, and shows how scientists actually measure it. It also covers the timescales involved, which stretch from a few hours to an entire human lifetime.

Persistence, Resistance, and Tolerance: The Core Distinction

Students lose the most points on this distinction, so start here.

Resistance is a heritable, genetic property. A resistant bacterium carries a mutation or an acquired gene that raises the minimum inhibitory concentration (MIC) of an antibiotic, meaning the drug no longer works at concentrations that used to kill it. Resistance is measured, it is stable, and it spreads to daughter cells and often to other bacteria.

Persistence is a phenotypic, non-heritable state. A persister cell is genetically identical to its drug-sensitive siblings. It survives antibiotic exposure because it has entered a dormant, low-metabolism state, not because it acquired a defense gene. When the antibiotic is removed, persisters wake up and their offspring are fully sensitive again. The MIC of a persister population does not change.

Tolerance sits between the two. A tolerant population is killed more slowly than a normal one, but its MIC is unchanged. Tolerance is often a stepping stone that allows resistance mutations to emerge under prolonged drug exposure.

The practical consequence is that a standard MIC test can miss persistence entirely. A culture can look fully susceptible in the lab and still relapse in a patient, because a tiny fraction of dormant cells survived and later regrew.

Persistence in Microbiology: Persister Cells and Biofilms

What Persister Cells Are

Persister cells are a small subpopulation of a genetically uniform bacterial culture that survives lethal concentrations of an antibiotic. They were first described in the 1940s, when researchers noticed that penicillin killed almost all staphylococci but left a tiny fraction alive that could regrow into a fully sensitive culture.

The key features:

They arise spontaneously, at frequencies that can range from roughly one in a thousand to one in a million cells depending on species, drug, and growth conditions.
They are dormant or near-dormant, with reduced translation, reduced ATP production, and active toxin-antitoxin (TA) modules that shut down growth.
They are not mutants. Sequencing a persister and its killed sibling shows identical genomes.
They revert. After the drug is removed, persisters resuscitate and produce a population with normal susceptibility.

Toxin-antitoxin systems are a well-studied mechanism behind this dormancy. A TA locus encodes a stable toxin that halts growth and an unstable antitoxin that neutralizes it. When the antitoxin is degraded faster than the toxin, growth stops. A survey of 100 clinical Escherichia coli isolates found TA loci in most strains: mazF in 75%, ccdB in 80%, relE in 81%, mqsR in 58%, and hipA in 51%, and the presence of these systems tracked with biofilm formation [1]. That study illustrates how widespread the molecular hardware for persistence is in ordinary clinical isolates.

The stringent stress response is a second major route. In Klebsiella pneumoniae, the regulator DksA drives the stringent response and is required for survival against membrane-targeting antibiotics, for robust biofilm formation, and for gastrointestinal colonization in a mouse model [2]. DksA also supports environmental survival and transmission by regulating RpoS. This is a clean example of how a single stress-response regulator links persistence inside a host to persistence in the outside world.

Biofilms: Persistence at the Community Level

A biofilm is a surface-attached community of microbes embedded in a self-produced matrix of extracellular polymeric substance (EPS), typically containing polysaccharides, proteins, and DNA. Biofilms create persistence through several stacked mechanisms:

Diffusion barrier. The EPS matrix and the cells themselves slow antibiotic penetration, so inner cells see sublethal concentrations.
Metabolic gradients. Oxygen and nutrients run out toward the base of the biofilm, pushing deep cells into the same dormant state as persisters.
Phenotypic heterogeneity. Different regions of one biofilm express different genes, so no single drug hits everything.
Physical protection from host defenses. Immune cells and antibodies struggle to penetrate and clear the structure.

Biofilm structure depends on the environment. In Vibrio parahaemolyticus, biofilms grown at the gas-liquid-wall interface reached 6.4 to 7.3 log₁₀ CFU/cm², while fully submerged biofilms reached 5.9 to 6.3 log₁₀ CFU/cm². The submerged biofilms produced significantly more protein and polysaccharide per cell and formed heterogeneous, tower-like structures, while interface biofilms were more uniform [3]. Same organism, different physical setting, different persistence architecture.

Biofilm recalcitrance is now understood as a combination of tolerance and persistence, and it is a major driver of the antimicrobial resistance burden because it creates persistent, difficult-to-treat infections [4]. The cyclic dinucleotide c-di-GMP is a central regulator of the switch from free-swimming to biofilm lifestyle. In Vibrio cholerae, a membrane-associated diguanylate cyclase with GGDEF and sensory domains was identified as a driver of c-di-GMP synthesis and biofilm persistence, and computational screening of 1,092 natural compounds flagged luteolin and sativanone as candidate inhibitors that bound the modeled protein with affinities of about -9 kcal/mol [5].

Chronic Infection and Intracellular Persistence

Some bacteria persist by hiding inside host cells. Staphylococcus felis, a coagulase-negative staphylococcus normally associated with cats, was recovered from bone biopsies of a diabetic foot osteomyelitis patient five months apart. Both isolates formed robust biofilms and persisted inside macrophages and osteoblasts, with the strongest intracellular survival in osteoblasts [6]. That is a textbook illustration of how an intracellular niche plus biofilm formation can produce a chronic infection that outlasts a course of antibiotics.

Prosthetic joint infections follow the same playbook. They are biofilm-driven, shaped by taxonomic diversity, spatial organization, host responses, and implant surface interactions, which is why culture alone gives a partial picture and why researchers now combine sequencing, transcriptomics, metabolomics, and advanced imaging to capture adaptive persistence across implant surfaces [7].

Viral Persistence and Latency

Viruses persist by a different route: they keep their genome inside host cells, often with little or no protein production. This is latency, and it is the viral version of dormancy.

Herpes simplex virus establishes latency in sensory neurons and reactivates periodically to cause cold sores or genital lesions. The viral genome persists as an episome, and reactivation is triggered by stress, UV light, fever, or immune suppression.
Epstein-Barr virus (EBV) establishes life-long latency in human B cells. A recent study identified an EBV-encoded small nucleolar RNA, v-snoRNA1, that directs 2'-O-methylation of host ribosomal RNA at 18S-C621 and 28S-U1760, impairing ribosomal maturation and translational fidelity. Deleting v-snoRNA1 increased protein synthesis and proliferation and reduced viral production, showing that EBV actively tunes host translation to balance persistence against lytic replication [8].
Human cytomegalovirus (HCMV) also establishes latency. One study found that the viral gene UL2 is not required for lytic replication, latency establishment, or reactivation in THP-1 monocytes or primary human hematopoietic progenitor cells, suggesting UL2 serves a function outside strict viral fitness [9].
HIV is the best-studied persistent virus in medicine. It integrates into host DNA and establishes reservoirs that survive decades of suppressive antiretroviral therapy. The reservoir is not one thing. Resting memory CD4+ T cells are the best-characterized cellular reservoir, but tissue microenvironments shape persistence too. In the rectal mucosa, myeloid cells including tissue-resident macrophages and dendritic cells may support HIV-1 persistence through longevity, resistance to apoptosis, metabolic adaptation, epigenetic regulation, and sequestration of virions in virus-containing compartments [10]. A separate line of work showed that the HIV viral microenvironment in tissues shares multiple features with certain tumor microenvironments, suggesting shared immunoregulatory and survival mechanisms [11].

Viral persistence also involves host-state remodeling. In virus-associated cancers, retained viral material can act through continued viral-product expression, latency, episomal or proviral genome maintenance, integration, or covalently closed circular DNA, and the host can retain a memory of the infection even after viral suppression or clearance [12].

Persistence in Pharmacology: Drug Persistence at Receptors and in Tissue

In pharmacology, persistence describes how long a drug's effect lasts, which is not the same as how long the drug is detectable in blood.

Three related concepts:

Receptor occupancy persistence. A drug can remain bound to its receptor long after plasma concentrations fall. This is why some drugs have a duration of action that outlasts their half-life. Irreversible inhibitors (for example, aspirin's acetylation of platelet cyclooxygenase) produce effects that persist for the lifespan of the target cell, which for platelets is about 7 to 10 days.

Tissue persistence. Lipophilic drugs accumulate in fat, muscle, or organ tissue and are released slowly. This creates a deep compartment that prolongs the effective half-life and can extend the dosing interval well beyond what plasma kinetics would predict.

Functional persistence. Some drug effects persist through downstream signaling cascades or epigenetic changes that outlive the drug itself. This is the pharmacological version of a host-state memory.

The clinical relevance is straightforward. If a drug's effect persists longer than its measured plasma half-life, stopping the drug does not immediately stop the effect, and starting a new drug that acts on the same pathway can produce unexpected additive toxicity. Pharmacologists therefore distinguish pharmacokinetics (what the body does to the drug) from pharmacodynamics (what the drug does to the body), and persistence is where the two meet.

The same logic applies to immunomodulation. JAK inhibitors, now repurposed for infectious diseases after baricitinib showed a survival benefit in hospitalized COVID-19 patients, modulate immune signaling in ways that can alter viral control and chronic inflammation. Their effects on viral persistence and immune restoration remain heterogeneous, which is exactly why the timing and duration of exposure matter [13].

Persistence in Ecology: Population Persistence and Source-Sink Dynamics

In ecology, persistence means a population remaining present in a location over time despite pressures that would drive it extinct.

Source-Sink Dynamics

A source population produces a surplus of individuals that disperse outward. A sink population receives immigrants but cannot sustain itself without them. A sink persists only because of continuous immigration from a source. Remove the source, and the sink collapses.

This matters for conservation because a population can look stable while quietly depending on an external subsidy. Counting individuals tells you the population is there. It does not tell you whether it can persist on its own.

Minimum Viable Population and Stochasticity

Population persistence depends on surviving several kinds of randomness:

Demographic stochasticity. Chance variation in births and deaths matters most in small populations.
Environmental stochasticity. Droughts, storms, and temperature swings affect everyone at once.
Genetic stochasticity. Inbreeding and loss of diversity reduce fitness over generations.
Catastrophes. Rare events that remove a large fraction of the population in one stroke.

A population persists when its growth rate, its spatial connectivity, and its genetic diversity are sufficient to absorb these shocks.

Microbial Ecology as a Persistence Model

The same framework applies to microbes. In marine and food-processing environments, Vibrio parahaemolyticus biofilms persist across temperature, oxygen, and surface conditions, with interface biofilms favored by oxygen diffusion and submerged biofilms favoring EPS production [3]. In the gut, K. pneumoniae uses the stringent response regulator DksA to colonize and persist, and the gastrointestinal tract acts as a reservoir from which the organism can spread and cause invasive disease [2]. These are ecological persistence problems solved with molecular tools.

Comparison Table: Persistence Across Domains

Domain	Definition	Mechanism	Example	Timescale
Microbiology (persister cells)	Small subpopulation of genetically identical bacteria surviving lethal antibiotic exposure	Dormancy, toxin-antitoxin modules, stringent response, reduced metabolism	E. coli persisters with hipA, mazF, relE loci [1]	Hours to days per cycle
Microbiology (biofilms)	Surface-attached community that resists clearance	EPS matrix, diffusion barrier, metabolic gradients, c-di-GMP signaling	V. cholerae biofilm driven by a diguanylate cyclase [5]; prosthetic joint infections [7]	Weeks to years
Microbiology (intracellular bacteria)	Bacteria surviving inside host cells	Intracellular niche, immune evasion, biofilm co-formation	S. felis in osteoblasts and macrophages [6]	Months
Virology (latency and reservoirs)	Viral genome maintained in host cells with minimal or no replication	Episomal or integrated genome, transcriptional silencing, tissue niches, immunometabolism	HIV reservoirs in CD4+ T cells and rectal myeloid cells [2,7]; EBV latency in B cells [8]	Years to lifetime
Pharmacology (drug persistence)	Drug effect outlasting plasma exposure	Receptor occupancy, tissue accumulation, downstream signaling	Irreversible enzyme inhibition, deep-compartment accumulation	Hours to weeks
Ecology (population persistence)	Population remaining present despite extinction pressure	Source-sink dynamics, dispersal, genetic diversity, stress tolerance	Sink populations sustained by immigration	Generations to centuries

How Persistence Is Measured

Different fields use different assays, and knowing the assay tells you what claim is actually supported.

Time-kill curves. A bacterial culture is exposed to a fixed antibiotic concentration and sampled over time. A biphasic curve, a steep drop followed by a plateau, is the classic persister signature. The plateau represents surviving dormant cells.

MIC testing. The standard method for resistance. Persisters do not shift the MIC, which is why MIC testing alone cannot detect them. This is a common source of confusion in clinical microbiology.

Biofilm assays. Microtiter plate assays quantify attached biomass, often with crystal violet. More advanced models use flow cells, Calgary biofilm devices, or submerged coupon systems that mimic real surfaces [3].

Latency reactivation assays. For HIV, researchers use cell lines such as ACH2, a well-characterized model of HIV-1 latency, and a two-color flow cytometry readout to screen latency-reversing agents (LRAs). Using this approach, PEP005 and CUDC-907 were identified as the most potent LRAs across multiple settings, with transcriptomic analysis confirming upregulation of viral RNA copies, and Tandutinib was identified as a novel candidate with appreciable latency-reversing activity [14]. The ACH2 model does not fully recapitulate in vivo latency, which is a limitation the authors acknowledge.

Spatial and single-cell methods. ImmunoPET/CT-guided spatial transcriptomics can localize rare foci of viral infection in tissue and define the viral microenvironment, which is how researchers showed that persistent and transient SIV reservoirs differ in their local immune context [11].

Longitudinal sampling. For chronic infections, persistence is documented by recovering the same organism or genome from the same patient over months. The S. felis case is a clean example: two isolates recovered five months apart from bone biopsies of the same patient, with whole-genome sequencing revealing strain-specific variation in virulence regulation, phage defense, and iron acquisition genes [6].

Why Persistence Matters Clinically and Scientifically

Persistence is the reason several major medical problems exist.

Relapse after apparently successful treatment. A course of antibiotics can clear the susceptible majority and leave persisters that regrow. This is a leading explanation for recurrent urinary tract infections, chronic wound infections, and prosthetic joint infections [7].

The HIV cure barrier. Latent reservoirs are refractory to antiretroviral therapy and can drive rapid viral rebound when treatment is interrupted. The "shock and kill" strategy aims to induce viral transcription with LRAs so infected cells become visible to the immune system, but the approach remains limited by reservoir heterogeneity and tissue sanctuaries [14].

Immunometabolism and comorbidity. HIV reprograms the metabolism of CD4 T cells and myeloid cells, and these metabolic states differ between initial infection, active replication, and latency. These alterations may contribute to the increased frequency and severity of comorbidities in people with HIV [15].

Post-translational control. Beyond transcriptional silencing and epigenetic repression, post-translational modifications of host and viral proteins act as dynamic switches regulating integration, transcription, immune recognition, and reservoir maintenance, which makes them attractive targets for reversing latency or destabilizing reservoirs [16].

Aging and reactivation. Latent viruses such as CMV can reactivate when host resilience declines. Age-associated immunosenescence, inflammaging, mitochondrial dysfunction, and epigenetic drift can destabilize the networks that normally keep CMV latent, creating self-reinforcing loops in which inflammation promotes reactivation and reactivation amplifies inflammation [17].

Immune checkpoint modulation. PD-1/PD-L1 inhibitors may reverse latency, enhance HIV-specific immune responses, and contribute to reservoir regulation, though effects on reservoir dynamics and immune restoration remain heterogeneous and safety considerations are unresolved [18].

Common Mistakes and Limitations

Treating persistence as resistance. This is the single most common error. Persisters are phenotypic and reversible. Resistance is genetic and heritable. A persister population has an unchanged MIC. If you describe a persister as resistant, you have mislabeled the biology and probably the treatment implication.

Assuming a negative culture means clearance. Biofilm and intracellular organisms can be present but not recovered by standard culture, which is why prosthetic joint infection research increasingly combines culture with sequencing, transcriptomics, and imaging [7].

Confusing viral latency with viral clearance. A patient with undetectable HIV RNA on therapy still has a reservoir. Undetectable is not absent [10].

Ignoring tissue compartments. Blood is easy to sample and often unrepresentative. HIV persistence is shaped by tissue microenvironments, including the rectal mucosa, that differ substantially from blood [2,18].

Overgeneralizing from cell line models. The ACH2 line is a useful and reproducible latency model, but it does not capture the full complexity of HIV latency in vivo [14]. Model limitations are a real constraint on how far findings can be extended.

Assuming ecological stability means self-sufficiency. A sink population can look stable while depending entirely on immigration. Persistence in the short term does not guarantee persistence without the source.

Forgetting that persistence is not always bad. Persistence mechanisms also underlie long-term immunity, memory B cells, and stable microbiomes. The term is descriptive, not a value judgment.

Individual cases, whether in a patient or a wild population, require context-specific professional assessment. The general principles here do not substitute for that.

Quick Review

Persistence means remaining present over time despite pressures that should remove the entity.
Persisters are phenotypic, not genetic. Their MIC does not change. Resistance is genetic and heritable.
Toxin-antitoxin modules and the stringent response are core molecular mechanisms of bacterial persistence.
Biofilms create persistence through matrix barriers, metabolic gradients, and phenotypic heterogeneity.
Viral persistence works through latency and reservoirs, with HIV and herpesviruses as the classic examples.
Pharmacological persistence is about effect duration, not drug detection, and depends on receptor occupancy and tissue compartments.
Ecological persistence depends on source-sink dynamics, dispersal, and genetic diversity.

Frequently Asked Questions

What is the difference between persistence and resistance?

Persistence is a reversible phenotypic state in which a subset of genetically identical cells survives drug exposure without any change in MIC. Resistance is a heritable genetic change that raises the MIC and is passed to daughter cells.

Are persister cells a type of antibiotic resistance?

No. Persisters are dormant cells that tolerate antibiotics without carrying resistance genes. When the drug is removed, they resuscitate and produce fully susceptible offspring.

What causes viral persistence?

Viral persistence arises from latency, in which the viral genome is maintained in host cells with minimal replication, and from tissue reservoirs that protect infected cells from immune clearance. HIV and herpesviruses are the standard examples.

How long can persistence last?

Timescales range from hours for a bacterial persister cycle to years or a lifetime for viral reservoirs such as HIV and EBV.

What is a source-sink dynamic in ecology?

A source population produces surplus individuals that disperse outward. A sink population cannot sustain itself and persists only through continued immigration from a source.

Can persistence be reversed?

Bacterial persistence is reversible when the antibiotic pressure is removed. Viral latency can be pharmacologically induced with latency-reversing agents, though current approaches remain limited.