Maria de Sousa, the immunologist who mapped where lymphocytes live

Maria de Sousa died in the early hours of Tuesday 14 April 2020 in the intensive care unit of the Hospital de São José in Lisbon. She had been admitted a week earlier with Covid-19, the disease caused by the new coronavirus, SARS-CoV-2.2 sources2 sources

That morning the University of Porto announced the death of its emeritus professor of immunology. The rector, António de Sousa Pereira, called it an enormous loss for the university. Henrique Cyrne Carvalho, director of ICBAS, the medical school where she had taught since 1985, was also her friend and doctor. He called her death one of the worst pieces of news of those days, and said they had spoken by phone every day until she was moved to intensive care. The university added that many of the scientists now at the front line of research on the virus had been her students.1 source1 source

Eleven days earlier, on 3 April, she had written a poem in English. After her death the publisher Quetzal Editores posted it in full on social media, in a Portuguese translation by the doctor and poet João Luís Barreto Guimarães. Her first two PhD students called it a touching farewell.2 sources2 sources

Lymphocytes are the white blood cells that recognise germs. In 1959 James Gowans showed that they do not stay put: in rats they leave the blood, pass through the lymph nodes and return to the blood through the lymph, over and over.1 source1 source

The thymus, a small gland in the chest, was still a mystery. In 1961 Jacques Miller reported that mice whose thymus was removed just after birth grew up with poorly developed lymphoid tissue, weak immune responses and an unusual tendency to catch infections. Something made in the thymus was needed to build an immune system.2 sources2 sources

Maria de Sousa graduated in medicine from the University of Lisbon in 1963, after choosing medicine over a career as a pianist. Years later she said she turned to research because, as a young doctor, she had little to offer patients.2 sources2 sources

In 1964 she left Portugal, then still under the Estado Novo, the dictatorship founded by António de Oliveira Salazar, with a scholarship from the Calouste Gulbenkian Foundation. Her destination was the Experimental Biology Laboratories of the Imperial Cancer Research Fund at Mill Hill, in London.4 sources4 sources

At Mill Hill she worked on Miller's puzzle with the immunologist Delphine Parrott and a third colleague, J. East. They removed the thymus from newborn mice of two strains and looked under the microscope at the spleen and the lymph nodes, the small bean-shaped glands where immune cells gather. The organs were not uniformly empty. Lymphocytes were missing from particular places.2 sources2 sources

In the spleen the depleted zone lay inside the lymphoid follicles, immediately around the small central arteries. In the lymph nodes it was the mid and deep cortex, the band below the round clusters of cells near the surface. The depletion showed as early as three weeks after the operation. The team called these places thymus-dependent areas.1 source1 source

They then labelled thymus cells from healthy mice with radioactive adenosine and injected them into the blood of other mice of the same strain. The labelled cells settled preferentially in the thymus-dependent areas. Spleen cells, labelled the same way, went there too but also to the edges of the follicles and the outer cortex of the nodes.1 source1 source

The paper appeared on 1 January 1966. A second, in Nature that December, followed the same areas in lymph nodes after the immune system was stimulated. Her students later wrote that with these papers the T and B cell zones had arrived, and that Parrott and de Sousa were the first to describe the thymus-dependent areas.3 sources3 sources

She moved to Scotland in 1967 or 1968 (Lusa gives the first year, the FCT the second) and took a teaching post at the University of Glasgow, where she also did her doctorate in immunology. The FCT and Lusa date the degree to 1972; her students give 1971.4 sources4 sources

A new animal helped. In January 1968 E. M. Pantelouris reported that a mouse mutant, the nude mouse, has no thymus. De Sousa, Parrott and Pantelouris examined seven of them against eight normal littermates. In every lymph node from the nude mice the thymus-dependent areas were virtually empty of lymphocytes, as if nature had done the thymectomy.2 sources2 sources

Then she asked how the cells find their places. Working in the Department of Immunology at Glasgow's Western Infirmary, she labelled mouse thymus cells and bone marrow cells and injected them into the blood of matched mice. Thymus cells went mostly to the thymus-dependent territory. Most marrow cells went to the thymus-independent sites, and different kinds of marrow cell picked out different spots within them.2 sources2 sources

The paper had one author. Earlier the same year she and Parrott had published a review setting out the idea in general terms: thymus-derived and marrow-derived lymphocytes look the same in the blood, yet they sort themselves into separate compartments of the lymph nodes, spleen and Peyer's patches in the gut.2 sources2 sources

Ecotaxis described the behaviour without explaining it. The mechanism came almost thirty years later, from mice engineered to lack single genes. Chemokines are chemical signals that attract cells, which detect them through receptors on their surface. Without the chemokine receptor CCR7, lymphocytes migrate poorly and every secondary lymphoid organ is badly disorganised. B cells need another receptor, CXCR5, and its chemical signal, BLC, to reach their follicles.2 sources2 sources

Her students, writing in 2020, called the 1971 paper visionary, and pointed out that it anticipated the cell niches that biologists now discuss for blood stem cells and plasma cells.1 source1 source

In 1975, according to Lusa, she crossed the Atlantic. In New York she was an associate professor at Cornell Medical College and directed the Cell Ecology Laboratory of the Sloan Kettering Institute, part of Memorial Sloan Kettering Cancer Center; she also worked at Harvard Medical School. Her first two PhD students, Antonio Freitas and Benedita Rocha, followed her from Glasgow; they worked with her from 1973 to 1977.5 sources5 sources

She tried the ecotaxis idea on human disease. A 1977 study of five children with Hodgkin's disease, a cancer of the lymphatic system, argued that a shortage of lymphocytes in the blood did not always mean the body lacked them, and that failed ecotaxis could happen in people. A year later she went further, proposing that iron-binding proteins such as ferritin, transferrin and lactoferrin help control where lymphocytes go.2 sources2 sources

That link between iron and the immune system became the second half of her career.2 sources2 sources

She returned to Portugal in 1984, interrupting a rising career in the United States, and in 1985 became professor of immunology at ICBAS, the University of Porto's biomedical sciences institute. There she set up a research programme on haemochromatosis, a hereditary illness in which the body stores too much iron. Her students note that it is particularly common in the north of Portugal.4 sources4 sources

With the haematologist Graça Porto she built a team spanning ICBAS, the Hospital Geral de Santo António and the Institute for Molecular and Cell Biology (IBMC). In 1989 they reported that the disease in northern Portugal was associated with a particular combination of tissue-type (HLA) genes, A3 and B7. HLA genes are the markers matched between donor and patient in transplants.3 sources3 sources

Results in mice and patients followed in 1994. The Porto group reported that mice lacking beta-2-microglobulin, a protein of the immune system, built up iron in the liver in a pattern indistinguishable from the human disease. In 21 patients treated by repeated bloodletting, the standard treatment, the team found that each person's balance of two kinds of T cell, CD4 and CD8, stayed stable over time and correlated with the amount of iron removed.2 sources2 sources

In August 1996 an American team found the human haemochromatosis gene, later called HFE. It belongs to the same family as the immune system's MHC class I genes, whose molecules are assembled with beta-2-microglobulin. That November a study from Utrecht, with de Sousa among the authors, showed that mice without beta-2-microglobulin fail to limit the passage of iron from the gut into the blood, and concluded that a gene product depending on it is involved in iron balance.4 sources4 sources

She also changed how Portuguese science was run. In 1987 José Mariano Gago, then president of JNICT, the national research board, and later science minister, asked her to coordinate the creation of research units in the health sciences. She chaired its committee for biomedical research and introduced external peer review of projects and institutions.2 sources2 sources

She then coordinated the health sciences in the first external and independent evaluation of Portuguese research centres. The FCT dates this to 1991, under JNICT; Lusa and RTP place it in the mid-1990s, when Gago was minister.3 sources3 sources

Her largest legacy was people. She created a master's programme in immunology, which her students date to 1984 and the University of Porto to her arrival at ICBAS in 1985. In 1996 it grew into GABBA, a doctoral programme in basic and applied biology run by ICBAS, the university's faculties of science and medicine, IBMC, INEB and IPATIMUP. Freitas and Rocha describe it as the first time a professor at a Portuguese university had set up a doctoral programme.2 sources2 sources

GABBA gave every student a four-year scholarship and let them choose their subject, supervisor and laboratory, in Portugal or abroad. She kept in touch with all of them, whatever their field. The University of Lisbon, awarding her its prize in 2017, said the two programmes had shaped the community of Portuguese researchers working at home and around the world.2 sources2 sources

She retired as a full professor on 16 October 2009 with a final lecture, Uma escola sem muros (A school without walls), in the great hall of the University of Porto's rectorate, and was later made an emeritus professor. She stayed on as an honorary researcher at i3S, the university's health research institute; its director said she had been with them for twenty years.3 sources3 sources

She also wrote. A pianist and poet, she published in 2014 a book of reflections on the history of biomedical research in Portugal, Meu dito meu escrito. In the first days of April 2020, as the pandemic spread, she wrote Love letter in a virus pandemic and dedicated it to the GABBA students. It ends with these words.4 sources4 sources

Portugal's first Covid-19 death appeared in the daily report of 17 March 2020. Four weeks later, the report of 14 April counted 567. Maria de Sousa had several other illnesses that already made walking difficult. The University of Porto, RTP and Lusa gave her age as 81; none of them gave her exact date of birth.4 sources4 sources

The tributes came quickly. On 15 April the Ordem dos Médicos, Portugal's medical association, created a research prize in her name for doctors who do research. On 14 July 2020 i3S named its library after her, and the science minister, Manuel Heitor, called her one of the pillars of modern science in Portugal. On the first anniversary of her death the FCT, the national research funder, announced a call for 50 doctoral fellowships in epidemiology and virology bearing her name.3 sources3 sources

The division of lymph nodes and spleen into B cell follicles and T cell zones began with Parrott, de Sousa and East's thymectomised mice in 1966, and the question she named in 1971, how a cell knows where it belongs, is still being answered gene by gene.4 sources4 sources

In Portugal her mark is mostly in people and habits: the scientists trained through the Porto master's and GABBA, and the external peer review she pushed for in research funding. Her students noted in 2020 that review was still not fully practised in every Portuguese university.2 sources2 sources