Marburg virus

• 1. Introduction

  ›

  In 1967, laboratory workers at the Behringwerke, a major industrial plant in the German city of Marburg, handled tissues from grivet monkeys. These animals are also called African green monkeys. That same year, a mysterious illness appeared in Marburg, in Frankfurt and in Belgrade, then the capital of Yugoslavia. Thirty-one people became infected, and seven of them died.

  The cause was a pathogen that science had never described before. It became known as Marburg virus, a member of the same family as Ebola, and it causes a hemorrhagic fever in primates.

  This documentary follows the virus from its odd, filament-like shape to the places where it has struck since. Where does it live when no one is sick? How does it break into a living cell? Why is there still no approved vaccine? And why did a Cold War superpower take an interest in it?

  1112282930

• 2. Anatomy of a Filament

  ›

  A single Marburg virus particle is about 80 nanometres wide, yet its length is far less predictable. Typical particles run from 795 to 828 nanometres, noticeably shorter than Ebola particles, which measure between 974 and 1,086. In tissue culture, some stretched as long as 14,000 nanometres. Their outlines are just as unruly. Like other filoviruses, they can curve into a shepherd's crook, a letter U or a numeral 6, and some are coiled, ring-shaped or branched.

  Taxonomists place the virus in the family Filoviridae, within the order Mononegavirales. The genus name Marburgvirus joins the name of the city of Marburg, in Hesse, to the standard suffix for viruses. A tangle followed in 2005, when the virus was renamed Lake Victoria marburgvirus. That left italics as the only way to tell the organism's name from the name of its whole species. Scientific papers mostly kept the old name anyway. In 2010, the name Marburgvirus was restored and the species name changed instead.

  Seven structural proteins build each particle. At the core sits the helical ribonucleocapsid, made of the genetic RNA wrapped around a polymer of nucleoprotein, called NP. Attached to it are the polymerase L, its cofactor VP35 and a transcription activator, VP30. Two matrix proteins surround this core, the major one VP40 and the minor one VP24. Outside everything lies a lipid membrane borrowed from the host cell. Anchored in that membrane is a glycoprotein, GP1,2, which juts out as spikes 7 to 10 nanometres long. The architecture is nearly identical to Ebola's, yet the two viruses are antigenically distinct.

  The genetic material is one unsegmented strand of RNA about 19 thousand base pairs long, with negative polarity. It carries seven genes in a fixed order: NP, VP35, VP40, GP, VP30, VP24 and finally L. Counting from the 3' end of the strand, NP comes first and L comes last, and that placement has consequences inside the cell.

  1013161925

• 3. Inside the Cell

  ›

  Two independent studies, reported in the same issue of the journal Nature, pointed to a single protein called NPC1. It is a cholesterol transporter, and Ebola virus needs it to enter and replicate in cells. Cells taken from patients who lacked NPC1 survived exposure to Ebola in the laboratory and appeared immune. The same studies found that Marburg virus depends on NPC1 as well.

  The contact is direct. NPC1 binds to the viral envelope glycoprotein, and the second lysosomal domain of NPC1 is the part that makes the connection. That raised a striking possibility. Genetic mutations in the human NPC1 gene might make some people resistant to one of the deadliest known viruses. One study showed that a small molecule could block Ebola infection by stopping the glycoprotein from binding NPC1. The other found that mice with only one working copy of the gene were protected from a lethal dose of mouse-adapted Ebola.

  Once inside, the life cycle starts with the virion attaching to receptors on the cell surface. The viral envelope then fuses with cellular membranes, and the nucleocapsid spills into the cytosol. There the viral polymerase partly uncoats the nucleocapsid and copies the genes into messenger RNAs, which the cell translates into proteins.

  The polymerase L binds a single promoter at the 3' end of the genome. After each gene, transcription either stops or carries on to the next one. Genes near the 3' end are therefore copied most often, and those near the 5' end least often. The gene order works as a simple but effective dial for regulating output.

  The nucleoprotein is the most abundant product, and its growing concentration tells L when to switch from transcribing genes to copying the whole genome. Full-length positive-stranded antigenomes are made first, and these in turn yield negative-stranded copies for new virions. New proteins and genomes assemble near the inside of the cell membrane. Virions then bud outward, wrapping themselves in a piece of the very membrane they leave.

  2122232442

• 4. Bats and Ancestry

  ›

  In 2009, researchers reported isolating infectious Marburg virus from healthy Egyptian fruit bats, Rousettus aegyptiacus, that had been caught in the wild. Infectious Ravn virus was isolated as well. Together these results strongly suggest that Old World fruit bats help keep marburgviruses circulating in nature. Whether the Egyptian rousettes are the true hosts, or merely pick up the viruses from another animal, still needs further study.

  A first experimental infection study in 2012 added detail. The infected bats developed a relatively low level of virus in the blood, lasting at least five days, and they stayed healthy with no notable gross pathology. The virus still multiplied to high levels in the liver and spleen. It also grew in organs that might spread it, including the lung, intestine, reproductive organs, salivary gland, kidney, bladder and mammary gland. The long stretch of virus in the blood might even let blood-sucking arthropods carry it mechanically, in addition to direct contact with infected blood.

  Genetic comparison splits the viral strains into two clades, Ravn virus and Marburg virus. The Marburg strains divide again into A and B. The A strains came from Uganda, including five from 1967, from Kenya in 1980, and from Angola in 2004-2005. The B strains came from the Democratic Republic of the Congo epidemic of 1999-2000 and from a group of Ugandan isolates collected in 2007-2009.

  One analysis put the average evolutionary rate of the whole genome at 3.3 times ten to the minus four substitutions per site per year. It dated the common ancestor of the Marburg strains to roughly 178 years before the study, with a wide credible range of 87 to 284 years. That points to the middle of the 19th century. The Ravn strains look far younger, with an origin about 34 years before the study, in the early 1980s. The most probable home of the Marburg ancestor was Uganda, while the Ravn ancestor most likely lived in Kenya.

  5426272563

• 5. The Outbreak Record

  ›

  Two viruses cause Marburg virus disease in people: Marburg virus itself and Ravn virus. Both qualify as the same species because their genomes differ from the prototype by less than 10 percent at the nucleotide level. The illness brings haemorrhage, fever and other symptoms similar to Ebola.

  The early outbreaks were small. In 1975, three cases in Rhodesia and Johannesburg, South Africa, included one death. Kenya reported two cases in 1980, one of them fatal. In 1987, a single Ravn virus case in Kenya also ended in death.

  Then the numbers jumped. In the Congolese towns of Durba and Watsa, between 1998 and 2000, 154 people fell ill and 128 died, a fatality rate of 83 percent. Marburg virus and Ravn virus circulated together there, and the deaths were never split between them. Angola's outbreak of 2004-2005 was larger still, with 374 cases and 329 deaths, a rate of 90 percent.

  Uganda appears again and again in the later record. Four cases in 2007, involving both viruses, produced one death. In 2008, a fatal case turned up in Uganda and the Netherlands. The year 2012 brought 18 cases and 9 deaths. One fatal case followed in 2014, and three people died in 2017, every case in that outbreak.

  Guinea's government detected a case in 2021 by testing samples from patients who had died on the 2nd of August 2021 in the southern prefecture of Gueckedou. That prefecture lies near the borders with Sierra Leone and Liberia. In Ghana in 2022, four cases and three deaths were recorded. Two cases were confirmed on the 17th of July 2022, and two more on the 27th of July.

  The years that followed were busy. Equatorial Guinea reported 25 cases and 11 deaths in February 2023. Tanzania recorded 9 cases and 6 deaths in March 2023. Rwanda counted 58 cases and 13 deaths in 2024, a fatality rate of 22 percent. As of the 21st of January 2025, Tanzania reported one confirmed case, while 24 suspected cases tested negative and eight more were left without a stated cause. The most recent entries come from Jinka in Ethiopia, where 14 cases and 9 deaths were recorded from November 2025 until January 2026.

  4145464862

• 6. Keeping It Contained

  ›

  Virus particles remain infectious in clotted blood for four to five days. Spread happens through the exchange of body fluids, and through smear or contact infection. Unprotected sex and broken skin are routes between people.

  Recovery does not always end the risk. In convalescent patients, the virus can persist in so-called immunologically privileged sites, where the immune system reaches poorly. Examples include the anterior chamber of the eye and seminal fluid, where the virus has been recovered several months after the acute illness faded. For that reason, survivors should be advised to use condoms and be provided with them.

  The wider toolkit is practical rather than high-tech. It includes effective identification of cases, cleaning, isolation, contact tracing and monitoring, protective clothing, safe waste disposal and safe funeral practices for people killed by the disease. According to the World Health Organization, early professional treatment of symptoms, such as rehydration, considerably improves the chances of survival.

  Official classifications reflect the danger. The World Health Organization rates the virus as a Risk Group 4 Pathogen, requiring containment equivalent to biosafety level 4. In the United States, the National Institute of Allergy and Infectious Diseases ranks it as a Category A Priority Pathogen. The Centers for Disease Control and Prevention lists it as a Category A Bioterrorism Agent. The Australia Group lists it as a biological agent subject to export control.

  167523

• 7. The Missing Vaccine

  ›

  In 2009, expanded clinical trials of a vaccine against both Ebola and Marburg began in Kampala, Uganda. Even so, the World Health Organization says no vaccine or antiviral treatment for Marburg has been approved.

  The first clinical study of a Marburg vaccine's efficacy took place in 2014. It tested a DNA vaccine, and the people inoculated developed some antibodies. Such vaccines were not expected to give definitive immunity, though. Animal work has used hamsters, mice and non-human primates. Mice are handy in the early phases but differ too much from humans, so primates are needed as well. Infection in macaques appears most similar to the human disease.

  Many approaches stumbled in those trials. Virus replicon particles worked in guinea pigs but lost their effect in primates. An inactivated virus vaccine proved ineffective. DNA vaccines showed some efficacy in primates, yet every inoculated animal still showed signs of infection. Some scientists have tried to build one injection that covers both Marburg and Ebola, which would be more practical and cheaper for developing countries. That combined shot caused no adverse reactogenicity compared with two separate vaccinations.

  A candidate called rVSV-MARV was developed by the Canadian government in the early 2000s, alongside vaccines for related Ebola viruses. Merck acquired rights to those candidates in 2014 but declined to pursue most of them, Marburg included, for economic reasons. Merck handed the rights back to the Public Health Agency of Canada. However, the vital production techniques Merck gained while bringing the rVSV-ZEBOV vaccine into commercial use in 2019, with GAVI funding, remain Merck's. One account holds that these legal monopolies block production and testing of rVSV-MARV.

  The broader explanation given is economic. Handling the virus at qualified facilities is costly, deaths are relatively few, and commercial interest is lacking. A vaccine has simply not come to fruition.

  On the 23rd of June 2022, researchers working with the Public Health Agency of Canada reported a promising result. Their recombinant vesicular stomatitis virus vaccine, named PHV01, gave guinea pigs a high level of protection when given about one month before infection.

  7786838780

• 8. A Cold War Weapon

  ›

  Three Soviet research institutes ran Marburg virus programs during the Cold War. One was the Virology Center of the Scientific Research Institute for Microbiology in Zagorsk, now called Sergiev Posad. Another was the Scientific-Production Association Vektor in Koltsovo, today the State Research Center of Virology and Biotechnology Vektor. The third was the Irkutsk Scientific-Research Anti-Plague Institute of Siberia and the Far East. All of this belonged to a sprawling Soviet biological weapons effort with both offensive and defensive branches.

  Because most of the work was highly classified, how far the program got remains unclear. The Soviet defector Ken Alibek claimed that a weapon filled with the virus was tested at the Stepnogorsk Scientific Experimental and Production Base, in what was then the Kazakh Soviet Socialist Republic and is now Kazakhstan. If true, development had reached an advanced stage. No independent source has confirmed it.

  The human cost surfaced in accidents. The outbreak record lists laboratory accidents at Koltsovo in 1988 and 1990, each with one death. Alibek was the first to describe in detail the death of the Koltsovo researcher Nikolai Ustinov, who died in at least one such accident.

  The legacy reaches into present-day law, because the virus is a select agent in the United States.

  8976904340