Timothy A. Springer

Adhesion molecules of the immune system

Springer began his research career in immunology by studying the molecular basis for cellular immunity using as an example the interaction of cytotoxic T lymphocytes with antigen-bearing target cells. None of the molecules involved in this interaction had yet been defined, including the long postulated antigen-specific T cell receptor. However, it was known that Mg²⁺ was required for immune cell-cell interactions, and that interaction between fibroblasts and the extracellular matrix was similarly Mg²⁺-dependent. Postulating that adhesion receptors were likely to be similar to antibodies and were unlikely to require magnesium, Springer hypothesized that other molecules must be required for antigen-specific interactions. He immunized animals with cytotoxic T lymphocytes and screened for monoclonal antibodies that blocked antigen-specific killing.^[17] Using this functional assay, his laboratory identified a set of lymphocyte function-associated (LFA) antigens required for cytotoxic T-cell-mediated killing and other T cell-dependent responses, including LFA-1 (CD11a/CD18), LFA-2 (CD2), and LFA-3 (CD58).^[18][19]

Springer's group then identified ICAM-1 as a ligand for LFA-1, demonstrating that this interaction is required for efficient antigen recognition by cytotoxic T cells, and later discovered ICAM-2 (CD102) and ICAM-3 (CD50), which have two to five immunoglobulin (Ig)-like domains and comprise a subfamily of the Ig superfamily.^{[20][21][22][23]} Using purified proteins on substrates including artificial lipid bilayers and flow-based adhesion assays, his lab discovered the first heterophilic (like-unlike) receptor–counter-receptor adhesion pairs in all of cell biology, including CD2–LFA-3 and LFA-1–ICAM.^[24][25] The LFA-1–ICAM interaction required Mg²⁺, and explained the Mg²⁺-dependence of cell adhesion in antigen recognition.^[26][27] Subsequently, Springer structurally resolved the LFA-1–ICAM-1 interaction, with Mg²⁺ at the ligand-binding interface.^[28]

Integrins

After joining Harvard Medical School, Springer discovered that one of the monoclonal antibodies he had created with Milstein was specific for a macrophage differentiation antigen he termed Mac-1.^[29] Remarkably, both Mac-1 and LFA-1 had alpha and beta subunits and their beta subunits migrated at apparently identical positions in SDS-PAGE. Cross-linking showed that each contained a single alpha and beta subunit that were non-covalently associated into heterodimers. Peptide mapping and immunological cross-reactivity showed that their beta subunits were identical and their alpha subunits were distinct. This work, published in 1982, was the first evidence for structural homology among molecules that would later be called the integrins.^[30] Knowing that LFA-1 was functionally important stimulated a search for a function for Mac-1, which was shown to be a receptor for the complement component iC3b (CR3), which had been previously defined functionally but not at the molecular level.^[31] Work with antibodies to the common β subunit led to the identification of yet another heterodimer with a distinct alpha subunit, termed αX.^[32][33] Thus, three heterodimers, αLβ, αMβ, and αXβ were defined. N-terminal sequencing of the αL and αM subunits showed that they were homologous, and thus had diverged from a common ancestral gene and constituted a family of related proteins.^[34] Together with previous evidence that they contained identical β-subunits, αLβ, αMβ, and αXβ, constituted a functionally important family of receptors that participated in cell-cell interactions.^[35][36]

Meanwhile, workers in the platelet, extracellular matrix, and cell surface fields were also working on multi-subunit receptors. The platelet receptor for fibrinogen contained two glycoprotein subunits termed IIb and IIIa.^[37] A monoclonal antibody to a cell surface receptor that blocked cell adhesion to laminin and fibronectin reacted with three distinct subunits in the same molecular weight range as αLβ, αMβ, αXβ, and IIb/IIIa.^[38] Proteins with high molecular weight, sharing a common subunit, were identified on lymphocytes and non-hematopoietic cells.^[39] After the discovery that an RGD motif in fibronectin is sufficient for recognition by its receptor and common to other extracellular proteins,^[40] the fibronectin and vitronectin receptors were isolated and their sequences determined^{[41][42][43][44]} and platelet protein IIb/IIIa was also shown to recognize RGD.^[45] Sequences of these receptors^[46][47][48] showed that their alpha subunit sequences were homologous to one another and to those earlier reported for LFA-1 (αLβ) and Mac-1 (αMβ).^[49] A large number of other papers appeared in 1986-1987 reporting further sequences and relationships among these receptors, including from the Springer lab on the β-subunit shared by LFA-1 (αLβ), Mac-1 (αMβ), and αXβ.^[50]

Richard Hynes chaired the 1987 Gordon Research Conference on Fibronectin.^[51][52] Hynes had worked on fibronectin, and in 1986 his group isolated a cDNA encoding a subunit recognized by an antibody to the laminin and fibronectin receptors, which he named integrin.^[53] He convened researchers from the diverse fields converging on this newly recognized family of receptors and proposed a unifying nomenclature.^[54][55] Springer described the first sequence relationships among family members and was the first to use an αβ nomenclature in which α subunits were named after their receptor, such as αL for LFA-1 and αM for Mac-1.^[56][57] However, because the meeting was focused on fibronectin, the fibronectin receptor was designated α5β1, conferring the β1 designation on the subunit that pairs with more α subunits than any other β subunit in the family. The ones discovered by Springer on white blood cells became the β2 integrins.^[58]

Contributions by Springer, Hynes, and Ruoslahti to the integrin field were recognized by the 2022 Lasker Basic Medical Research Award.^[59]

Integrin inside-out signaling

Immunologists had initially pushed back against the idea that the adhesion molecules discovered by Springer could contribute to antigen-specific recognition, as they were thought to prevent antigen-specificity.^[60] In a major conceptual advance, Dustin and Springer discovered that adhesiveness of LFA-1 on T cells is regulated. They showed that LFA-1 is inactive until other cell surface receptors are stimulated. The initial demonstration was with T cell antigen receptor stimulation and pharmacologic stimulation of protein kinases.^[61] Intracellular tyrosine kinases are downstream from the T cell receptor and receptor tyrosine kinases and G protein coupled receptors are now known to be equally effective.^[62] The process was termed “inside-out signaling” and explained how adhesion receptors can cooperate with and amplify antigen recognition without abolishing immune specificity.^[63] It occurred without any change in LFA-1 density on the cell surface, suggesting that conformational change might be involved, stimulating Springer to become a structural biologist.^[64] Some thought that binding of adaptors such as talin was sufficient for inside-out signaling.^[65] However, force transmitted by the actin cytoskeleton through adaptors to integrins that is resisted by integrin-bound ligands embedded in the extracellular environment is also required and essential for ultrasensitivity.^[66] Furthermore, all stimulators of integrin inside-out signaling stimulate actin polymerization.^[67] Integrin inside-out signaling in leukocytes and platelets correlates with large increases in actin polymerization and cellular shape changes upon activation.^[68][69][70] Most recently, ligand binding to integrins was shown to induce full conformational change within milliseconds, and to be the first step in integrin inside-out signaling.^[71]

LFA-1 and LFA-3–based therapeutics for autoimmune disease

Although Springer made the first antibodies to LFA-1 and LFA-3,^[72][73][74] he did not patent them. An antibody isolated by Hildreth and McMichael^[75] was licensed to Genentech, which was humanized^[76] and approved by the FDA in 2003 for moderate to severe psoriasis as Raptiva.^[77] Its generic name efalizumab simulate the letters F and L in LFA-1.

Springer and lab member Mike Dustin collaborated with Barbara Wallner at Biogen to clone the cDNA for LFA-3.^[78][79] The LFA-3 ectodomain, which contains one immunoglobulin-like domain, was fused to the Fc domain of IgG to create Amevive.^[80][81] Amevive was approved for moderate to severe psoriasis in 2003.^[80] Its generic name, alefacept, is a homophone of LFA plus "cept" from receptor. Royalties paid by Biogen to Dana Farber Cancer Institute were equally shared by Dustin and Springer.^[82][83]

Leukocyte adhesion deficiency and the "multistep paradigm"

Working with physicians seeing patients with recurring, life-threatening bacterial infections, Springer found their leukocytes lacked LFA-1 (αLβ2), Mac-1 (αMβ2), and αXβ2.^[84][85][86] Anderson and Springer named the disease leukocyte adhesion deficiency (LAD) in a review article,^[87] and it was shown to be caused by mutations in the β2 subunit common to the leukocyte integrins.^[88] Patients have abnormally high levels of neutrophils in their circulation, which cannot emigrate out of the bloodstream to fight infection.^[87] This demonstrated that β2 integrins were important in leukocyte interactions with vascular endothelial cells and stimulated Springer to set up work in the lab on endothelial cells and flow chambers.^[89] In related work on LFA-1 ligands, ICAM-1 and ICAM-2 were shown to be inducible and constitutively expressed, respectively, on endothelium.^[90][91][92]

It had long been known from intravital microscopy that leukocyte emigration involved leukocyte rolling on endothelium followed by firm adhesion and subsequent transendothelial migration. Using white blood cells infused in flow chambers and purified adhesion molecules on the chamber walls under physiological shear rates found in vivo, Springer and postdoctoral fellow Mike Lawrence reconstituted three sequential interactions required for leukocyte emigration.^[93] Flow chamber walls were coated with P-selectin, ICAM-1, or both. Infused neutrophils were found to readily attach and then roll on P-selectin, but could not attach to ICAM-1 in flow. With both P-selectin and ICAM-1 on the substrate, cells attached and rolled, but the presence of ICAM-1 had no effect. However, if a neutrophil chemoattractant was added to the flow stream, it activated GPCR signaling, actin polymerization, and integrin-dependent firm adhesion to ICAM-1.^[93][89] Springer integrated these findings into the “three step paradigm” of leukocyte extravasation (diapedesis), which has become the standard framework for understanding leukocyte trafficking in inflammation.^[89]

The laboratory of Eugene Butcher, including postdoctoral fellow Uli von Andrian, was working on emigration of leukocytes in vivo at the same time, and discovered that antibody to leukocyte integrins inhibited firm adhesion, but not rolling adhesion, to postcapillary venules at sites of inflammation.^[94] Butcher and Springer received the 2004 Crafoord Prize for this work.^[95]

Springer's paradigm that integrins on leukocytes bind to counter-receptors with Ig-like domains on endothelium was later extended by others to integrin α4β1 binding to vascular cell adhesion molecule-1 (VCAM-1), which is inducible by inflammatory mediators on endothelium, and α4β7 binding to mucosal addressin cell adhesion molecule-1 (MAdCAM-1), which is constitutively expressed on mucosal endothelium.^[96][97][98]

At the time that Springer described the three step model, no chemoattractants for peripheral blood lymphocytes that could stimulate emigration were known.^[89] Their existence was implied by the ability of pertussis toxin, which modifies the G protein Gαi subunit, to cause lymphocytosis in patients with whooping cough.^[89] Springer's lab screened for such activity in cell line supernatants, purified and sequenced the protein, and found that stromal derived factor (SDF-1), previously defined as a growth factor for B cells, was a potent chemoattractant for both B and T cells.^[99] SDF-1 activated an orphan GPCR, later named CXCR4, which was also the co-receptor for T-cell-tropic HIV; SDF-1 further blocked infection of T cells by HIV.^[100] SDF-1 (CXCL12) is also a chemoattractant for CD34+ hematopoietic stem cells and regulates their movement from bone marrow to the bloodstream.^[101] Based on these discoveries, plerixafor (Mozobil) was developed as an antagonist of CXCR4 and is approved, in combination with filgrastim, for use in mobilizing hematopoietic stem cells in patients with multiple myeloma or non-Hodgkin lymphoma.^[102]

In later retrospective discussion, Springer framed the three steps in leukocyte emigration into inflammatory sites as an “area code” model, emphasizing that each step requires a cognate receptor–ligand interaction and therefore provides multiple intervention points (selectins and ligands; GPCRs and ligands; integrins including LFA-1, α4β1, α4β7 and endothelial ligands). He argued that the size of the target space exceeded what could be pursued in an academic laboratory and helped motivate company formation.^[83]

LeukoSite

Springer has played a prominent role in biotechnology. His scientific and entrepreneurial work has contributed to multiple FDA-approved therapies in immunology, oncology, and rare diseases, and he has been an influential early investor and company founder in the life sciences industry.^[83][103]

In the early 1990s, Springer founded LeukoSite to develop therapeutics based on leukocyte adhesion and trafficking biology.^[83] LeukoSite’s programs advanced or incorporated three agents that ultimately became approved medicines: vedolizumab (Entyvio), a humanized antibody targeting integrin α4β7 for inflammatory bowel disease; bortezomib (Velcade) for multiple myeloma and mantle cell lymphoma; and alemtuzumab (Campath-1/Lemtrada) for B-cell chronic lymphocytic leukemia and multiple sclerosis.^{[83][104][105][106][107][108]} Millennium Pharmaceuticals acquired LeukoSite in late 1999 by issuing shares; after the merger, LeukoSite shareholders owned 35% of Millennium stock.^[109][83]

LeukoSite’s antibody program targeting integrin α4β7 was the foundation for vedolizumab (Entyvio), approved by the FDA in 2014 for ulcerative colitis and Crohn’s disease.^[83][110] Entyvio has since become a major product in the immunology space: Takeda, which acquired the asset through its purchase of Millennium, reported Entyvio revenue of JPY 914.1 billion, approximately US$6.1 billion, for the fiscal year ended March 31, 2025.^[111][112] More than 350,000 patients have been treated with Entyvio since 2014 in the U.S., based on Symphony claims data (June 2014-Jan 2025). Takeda has described Entyvio as a cornerstone of its portfolio.^[113]

Entrepreneurship and biotechnology investing

Springer co-founded biotechnology companies Scholar Rock in 2012,^[114] Morphic Therapeutic in 2015,^[115] Tectonic Therapeutic in 2019,^[116] and Seismic Therapeutic in 2022.^[117] He was also an early investor in Selecta Biosciences and Editas Medicine.^[118]

Springer was a founding investor of Moderna after investing USD$5 million in 2010.^[119] He was the company's fourth-largest shareholder and made USD$400 million when the company launched its initial public offering (IPO) in 2018.^[120][121] During the COVID-19 pandemic, Forbes estimated Springer's net worth as USD$1 billion after in share price of biotechnology companies surged.^[122]