Introduction

Measles is estimated to have diverged from rinderpest, a devastating cattle pathogen, around the sixth century BCE [1]. This is thought to have coincided with the rise of large cities, allowing for populations large enough (~250,000–500,000 people) to support continuous measles virus transmission. Measles today is highly adapted to humans and despite its zoonotic origin, there are no animal reservoirs for measles. Measles is the most contagious human virus and has co-existed with us for hundreds, if not thousands, of years. Nonimmune individuals exposed to measles have a 90% likelihood of infection. The term measles was coined in 1693 by Thomas Sydenham, from the medieval English mesles and the Latin misella, a diminutive of misery [2]. Measles was endemic in much of Europe, Asia, India, and China since the Middle Ages. Following European colonial expansion in the 16^th century, measles spread worldwide with devastating consequences. Severe measles epidemics in Cuba and Honduras reportedly killed nearly 67% of the total native population from 1529-1531 [2,3]. In the wake of a resurgence of measles cases worldwide, it is of interest to note some unique paradoxes of this virus.

1. Despite being an RNA virus with an error‑prone polymerase, measles remains antigenically stable over time

The measles vaccine originated in 1954 when a measles outbreak occurred at a boarding school in Massachusetts. Dr. Thomas Peebles and John Franklin Enders successfully cultivated the virus from 11-year-old David Edmonston, termed the “Edmonston-B” strain [4]. The Edmonston-B strain was attenuated by in vitro passage in human and chicken cells to generate the measles vaccine which was licensed for public use in 1963 [5]. The vaccine strain is most closely related to genotype A viruses, which are extinct and are not associated with documented endemic transmission in any part of the world [6]. This unique feature allows for distinguishment between natural infection and a potential vaccine-induced adverse event. While the vaccine is a live-attenuated virus, no human-to-human transmission of the measles vaccine strain has ever been documented [7]. Two doses of the measles vaccine confer 97% protection against disease.

Unlike influenza, measles virus only has one serotype or “monotype”, and therefore the antibodies generated in response to infection recognize conserved epitopes across all genotypes [7]. The measles genotypes are determined by the sequence of 450 nucleotides at the end of the nucleoprotein-coding region (termed the N-450 sequence) [8]. In 2003, there were 18 co-circulating genotypes worldwide. This number has since decreased steadily to only 2 genotypes, B3 and D8, since 2021 [8]. In the US and Canada in 2025, the outbreak strain was genotype D8, with only a minority of cases being from genotype B3 [9]. These B3 cases are suspected to stem from separate imports and not due to local transmission, as B3 strains have been circulating in regions of Africa and the Middle East [10]. During this outbreak, the World Health Organization reported that there is “no evidence of decreased vaccine effectiveness or changes in the virus that would result in increased severity” [9].

How is it possible that measles circulates in the global population under selective pressure by the measles vaccine, and yet we don’t detect changes in the virus? Measles is a negative sense RNA virus and requires an RNA-dependent RNA polymerase in order to transcribe and replicate its genome. This process is notoriously error prone, with an estimated error rate of 10⁻⁴–10⁻⁵ per site per replication [11]. However, field isolates during outbreaks show very low levels of nucleotide divergence suggesting that measles virus populations do not evolve rapidly [11–13].

What is keeping measles virus from mutating? Measles virus has high genetic stability observed both in laboratory settings and in field isolates. The most variable region of measles virus is the aforementioned N-450 sequence, but even this sequence is stable during outbreaks. One possible contributing factor is the “rule of six” that constrains paramyxovirinae family members to a genome length that is a multiple of six [14]. These six nucleotides are the length required to wrap around one nucleoprotein molecule, and this organization would not favor insertion or deletion mutations. Pneumovirinae, such as RSV, have similar genetic makeup but are not constrained by the “rule of six”, and these viruses display comparatively high genome variability. Thus, this rule of six may be one possible reason why measles virus exhibits robust genome stability compared to other members of the paramyxovirus family.

In addition to the “rule of six”, experiments performed by Munoz-Alia et al. suggest that measles virus is also constrained by B-cell epitopes on measles virus surface glycoproteins [15]. Hemagglutinin (H), the measles virus glycoprotein responsible for receptor binding, contains seven known antigenic regions, or locations in which antibodies can be made to neutralize virus infectivity. Some of these antigenic regions overlap surfaces necessary for binding viral entry receptors including the epithelial receptor, nectin-4, and the lymphocyte receptor, Signaling Lymphocytic Activation Molecule (SLAM). These measles receptors will be discussed further in paradoxes #2 and #3 below. At least five of these seven sites needed to be disrupted in order to affect the susceptibility of measles virus H to serum neutralization, and therefore conferred no selective advantage to the virus [15]. When these antigenic sites were mutated to escape serum neutralization, the virus lost its ability to bind one of, if not both, receptors [15]. In addition, even if measles virus H is not sensitive to serum neutralization, anti-measles virus Fusion (F) antibodies could still effectively neutralize the virus. However, a virus lacking tropism to SLAM and nectin-4 would prevent pathogenicity and transmission [16,17]. Munoz-Alia and colleagues concluded that “there is a near zero probability for the accidental emergence of a pathogenic measles virus capable of evading vaccine-induced immunity” [15].

To date, there is no evidence of measles vaccine-escape mutants. This is good news, as it supports the continued efficacy of the current vaccine to prevent future measles outbreaks, if widely utilized. The current measles vaccine is given as part of a combined measles-mumps-rubella (MMR) vaccine in a 2-dose series to achieve >95% immunity. The recent resurgence of measles has highlighted the importance of widespread use of vaccination and public health efforts to rebuild confidence in vaccines and increase vaccine adherence.

2. Measles virus infection is immunosuppressive but leads to lifelong immunity against itself

Multiple epidemiological studies document that children with a history of measles are more likely to be hospitalized or die from other infectious diseases up to 5 years after clearance of the initial infection [18–20]. Infection by measles virus leads to an induced immunosuppression called immune amnesia, a condition in which the immune system “forgets” previously acquired memory against other pathogens.

Immune amnesia is thought to occur through the targeting of multiple important cell types for the immune system. Recent studies in experimentally infected macaques and observational studies in humans found measles infection resulted in up to a 73% reduction in the preexisting repertoire of pathogen-specific antibodies [21,22]. This reduction took patients 2–3 years to recover protective immunity against non-measles pathogens. There are currently no medical guidelines to help protect patients recovering from immune amnesia, but it has been suggested revaccination with routine childhood vaccines may help mitigate secondary infections.

But how does this occur? One of the primary receptors measles virus utilizes is SLAM (aka CD150), a surface glycoprotein expressed not only on naïve and memory B cells and T cells, but also macrophages, mature dendritic cells, and plasma cells in lymphoid tissues and peripheral blood [23,24]. Measles preferentially infects T and B cells, which leads to subsequent immune-mediated clearance of SLAM+ lymphocyte populations [25,26]. Immune amnesia coincides with lifelong immunity against measles. While preexisting memory T and B cells are depleted, this lymphocyte depletion is masked by a rapid proliferation of new measles-specific lymphocytes [25,27]. These population changes in lymphocyte profiles remain detectable more than a month after recovery from initial infection.

But how do we know protection from measles infection is lifelong and why is this immune response so protective? Measles infection generating lifelong immunity was reported by Peter Ludwig Panum when he observed a measles outbreak on the Faroe Islands in 1846. He noted that individuals who had had measles when the previous 1781 outbreak had occurred on the island, were exempt from measles infection during this 1846 outbreak, even though 65 years had passed [28].

Measles infection leads to a durable immune response though multiple potential mechanisms occurring both during and in the months following infection. Because SLAM⁺ antigen-presenting cells are directly and productively infected, this may play a role in antigen presentation and B cell maturation. Infection of these antigen-presenting cells could lead to robust priming of the innate immune response, which would in turn lead to generation of high affinity measles-specific B cells. Further experimental evidence is needed to validate this hypothesis.

Measles virus-infected lymphocytes and dendritic cells can induce apoptosis. Furthermore, virus-infected monocytes can also induce apoptosis in neighboring uninfected lymphocytes by upregulation of apoptosis associated molecules, such as CD95 and TNF-related apoptosis-inducing ligand-receptor (TRAIL-R) [29–31]. Together, these processes likely contribute to the lymphopenia and immunosuppression observed during infection, however apoptosis alone does not account for all aspects of lymphocyte impairment during immune amnesia.

Another feature of measles that may contribute to robust protection is that viral RNA stays in the blood and tissues weeks to months after clearance of the infectious virus [32–34]. As a result, the immune system remains active for months. Both CD4 and CD8 T cells are needed to clear the virus, but both T cells and antibodies are necessary to reduce viral RNA [35]. A cellular response is necessary for initial virus control, and a humoral response is necessary for clearance of viral RNA and suppression of recurrent infectious virus replication/production. It is speculated that lifelong protective immunity after primary measles virus infection could be a consequence of prolonged antigen stimulation [33,36]. Immune activation and lymphocyte proliferation are evident for months after resolution of rash, and during this period, there is a shift in cytokine production towards promoting B cell proliferation and production of antibody-secreting cells [34].

But what about the vaccine? The measles vaccine induces both antibody and cellular immune responses. Protection is best correlated with the quality and quantity of neutralizing antibodies, primarily against H but with some contribution against F [34]. Antibodies alone protect from disease/rash but not from infection. T cells alone do not protect from infection or disease but facilitate clearance of viral RNA [35,37]. Over time antibodies levels and CD4 T cells decrease leading to a vaccine failure rate of ~5% [37–39].

It is important to note that unlike naturally acquired measles immunity, the measles vaccine does not result in immune amnesia [22]. By preserving the immunity of children infected within the first 10 years of life, the measles vaccine may have drastically lowered the baseline childhood morbidity and mortality rate [18].

3. Measles is a respiratory virus where the respiratory epithelia are not the initial site of infection

We know of measles as a respiratory virus due to the obvious respiratory symptoms and transmission. Often, a respiratory virus enters a host via the upper respiratory system and exits via that same respiratory route. Previously, Fields Virology 1^st–5^th editions cited that the “initial infection [of measles virus] is established in the respiratory tract” [40,41]. As a result of a series of studies, this paradigm was rewritten in the 6^th edition [42]. It is now accepted that measles instead infects tissue resident SLAM⁺ lymphocytes, dendritic cells, or macrophages in the respiratory epithelia first, these migratory SLAM⁺ cells travel to the lymph nodes, closely followed by systemic spread (Fig 1, upper panel) [42]. This systemic spread occurs during the prodromal phase, or period before the appearance of the characteristic measles rash. Finally, the virus is brought back to the respiratory epithelia via lymphocytes, macrophages, and/or dendritic cells, leading to infection and finally viral shedding (Fig 1, lower panel) [43–46]. Interestingly, systemic infection of the immune cells occurs asymptomatically. The diagnostic onset of rash and respiratory symptoms instead coincides with return of virus to the epithelium.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Graphical Summary of Paradox #3.

Measles virus enters the host via SLAM⁺ lymphocytes and myeloid cells, such as dendritic cells or pulmonary macrophages, then travels via lymphatic vessels for systemic spread. Measles virus is potentially returned back to the respiratory tract via migratory lymphocytes, macrophages, or dendritic cells traveling in blood vessels (for simplicity, only dendritic cells are depicted in the figure). SLAM⁺ cells bring virus to interact with nectin-4+ airway epithelial cells from the basolateral side. Virus then spreads laterally until infected cells are shed into the lumen to be coughed out, leading to transmission. Created in BioRender. McCray, P. (2026). https://BioRender.com/vu62axw.

https://doi.org/10.1371/journal.ppat.1014135.g001

The epithelial receptor to measles is nectin-4 (aka PVRL4) [47,48]. Nectin-4 is an adherens junction protein, functioning to maintain the structural integrity between airway epithelia. Infection in the lung is mediated by nectin-4, which is located at the basolateral side of airway epithelial cells [49]. Infection from the apical side facing the lumen of the lung is prevented by tight junctions between cells that restrict receptor access to the virus [49]. However, the receptor is available from the basolateral side, further supporting the model that infection of the lung occurs subsequent to infected migratory lymphocytes moving to the lungs.

Once an airway cell is infected, measles virus spreads directly from one infected cell to its uninfected adjacent neighboring cell [50]. This process continues to form an infectious center which can include anywhere from 30 to 100+ cells. In primary human airway epithelia cells, these infectious centers are able to detach as a single unit and the majority of virus is cell-associated within these infectious centers, as opposed to free virus on the cell surface or in the basolateral media [46]. Observational evidence of infectious centers has been documented from macaque infections, where multinucleated cells or infection associated lesions are present in the epithelium [26,45]. High levels of epithelial cell infection and epithelial damage due to shedding of infectious centers into the lumen may instigate coughing and sneezing, leading to virus transmission. These transmission events may include the release of both free measles virus as well as measles-infected cells or cellular debris.

Further evidence supporting the lung as the final site of infection was highlighted by a study in which rhesus macaques were experimentally infected with a nectin-4-blind measles virus. The resulting symptoms did not include any respiratory symptoms. Instead, macaques only exhibited malaise and skin rash, known symptoms of systemic viral infection. In addition, the humoral immune response resembled wild-type measles virus-infected animals and the macaques still produced strong neutralizing antibodies and controlled viral replication [16]. However, there was no detectable shed virus from the airways. Conversely, when macaques were infected by a SLAM-blind measles virus, there were no clinical signs of infection, including skin rash or anorexia [17]. Taken together, these experiments are consistent with the updated paradigm that infection of airway epithelial cells is not necessary for initiation of systemic infection, however it is necessary for host-to-host viral transmission.

Conclusion

In summary, measles virus offers a series of a seemingly self-contradictory observations that prove to be well founded. These paradoxes challenge assumptions about viral evolution, immunity, and pathogenesis. Despite being an RNA virus with an inherently error‑prone polymerase, measles remains antigenically stable, constrained by the structural and functional requirements of its glycoproteins. Although natural infection induces profound immune amnesia, it generates lifelong immunity against itself. While measles is transmitted as a respiratory pathogen, its replication strategy places the airway epithelium at the end, rather than the beginning, of the infection cycle. Together, these features underscore the extraordinary evolutionary balance measles virus has achieved over centuries of coexistence with humans. While this review highlights some potential explanations for these paradoxes, there are still many open questions regarding measles biology and pathogenesis. Understanding these biological contradictions not only provides insight into measles pathobiology but also highlights the necessity of widespread vaccination in preventing both acute disease and its long‑term immunological consequences.

Acknowledgments

We thank Angela Liu, Justin Kaufman, and Dr. Brajesh Singh for their critical reading of the manuscript.