Purified protein derivative of Mycobacterium tuberculosis and excretory-secretory antigen(s) of Toxocara canis expand in vitro human T cells with stable and opposite (type 1 T helper or type 2 T helper) profile of cytokine production.

Two infections. One body. Two immune responses that are mirror opposites of each other. The same person, carrying memory T cells shaped by tuberculosis bacteria and by a dog roundworm, produces entirely different cytokine signatures depending on which antigen you present to them. One set of cells makes interleukin two and interferon-gamma. The other makes interleukin four and interleukin five. Neither set ever crosses over. That clear opposition was the finding. The reason it mattered is that, until this work, a result like this had only ever been shown in mice. By the early 1990s, the Th1 and Th2 framework was one of immunology's sharpest ideas — in rodents. Mouse CD4-positive helper T cells had been shown to segregate into two largely mutually exclusive subsets defined by the cytokines they produce. Type one T helper cells, or Th1 cells, make interleukin two, interferon-gamma, and tumor necrosis factor-beta. Type two T helper cells, or Th2 cells, make interleukin four, interleukin five, interleukin six, and interleukin ten. Those cytokine differences map onto distinct immune outcomes: interferon-gamma is essential for delayed-type hypersensitivity, the skin-test reaction to tuberculin, while interleukin four and interleukin five drive IgE antibody production and eosinophilia — the hallmarks of allergic and parasitic responses. In humans, though, the picture was messier. Most peripheral blood CD4-positive clones looked like an intermediate state, producing a mixed cytokine profile. Clear Th1 or Th2 clones had been isolated, but mostly from tissues of patients in specific disease situations, not from ordinary circulating blood in healthy people. The question Del Prete and colleagues set out to answer was pointed: could natural infection stamp a stable, one-sided cytokine program into human T cells? Their antigen choices were deliberate. Purified protein derivative, or PPD, is the classical extract of Mycobacterium tuberculosis used in skin testing. Its whole clinical purpose is to reveal Th1-driven delayed-type hypersensitivity. Toxocara canis excretory-secretory antigen, or TES, comes from the roundworm that dogs carry. Human exposure to Toxocara is linked to elevated IgE and eosinophilia, the hallmarks of Th2 activity. PPD and TES were thus chosen as functional opposites, each one's known immunological footprint pointing in a different direction. The team recruited two healthy donors, aged 29 and 42, who had never had overt tuberculosis or toxocariasis but who showed a positive PPD skin reaction and whose blood cells proliferated in response to TES in culture. That double exposure — to a bacterium on one side and a helminth on the other — made these donors ideal for asking whether the same immune system had been tuned differently by different pathogens. From each donor, the team expanded antigen-specific T cell lines, then cloned them under limiting-dilution conditions — three hundred cells per well — so that each growing culture derived from a single cell. The resulting T cell clones are genetically uniform populations descended from one founding cell. Any cytokine result reflects that cell's committed program, not a mixture of tendencies across a crowd. Southern blot analysis of T cell receptor beta-chain gene rearrangements confirmed independent clonal origins across eight PPD clones and eight TES clones tested. At scale, the cloning yielded sixty antigen-responsive PPD-specific clones and sixty-nine TES-specific clones. The results were about as clean as biology gets. When each clone was stimulated with its cognate antigen, the split was near-total. Virtually all PPD-specific clones secreted interleukin two and interferon-gamma and not interleukin four or interleukin five. Most TES-specific clones secreted interleukin four and interleukin five and not interleukin two or interferon-gamma. Only four TES clones additionally produced interferon-gamma, giving them a mixed Th0-like pattern, and only two PPD clones deviated from the Th1 profile. The division held under a completely different kind of stimulus as well. When clones were hit with phorbol myristate acetate, plus anti-CD3 antibody, a maximal non-specific activation that bypasses antigen entirely, seventy-seven percent of PPD-specific clones retained their Th1 profile and twenty-three percent looked Th0. Not one was Th2. Among TES-specific clones, sixty-two percent were Th2 and thirty-eight percent were Th0. Not one was Th1. That persistence under polyclonal stimulation is critical. It means the cytokine program is not just a response to what the cell last encountered — it is intrinsic to the cell itself. The molecular data confirmed it. Slot-blot RNA analysis of randomly selected clones showed that TES-specific clones which failed to secrete interferon-gamma also lacked detectable messenger RNA for interferon-gamma. PPD-specific clones that produced no interleukin four or interleukin five had no transcripts for those cytokines either. The gene isn't being expressed at a low level and failing to produce protein — it is silenced. That transcriptional shut-down is the signature of a committed cell, not an undecided one. Then Del Prete and colleagues tested whether these phenotypes would last. Over roughly six months, they repeatedly assessed the cytokine profiles of fifteen Th1-like and twelve Th2-like clones. None shifted. A Th2 clone stayed Th2. A Th1 clone stayed Th1. Even under the strong polyclonal stimulus, some clones moved toward the intermediate Th0 state, but the transition was always toward the middle, never across it. No Th1 clone became Th2, and no Th2 clone became Th1. The commitment, once made, held for the full length of the study. What determines which way a naive T cell goes? Del Prete and colleagues raise three candidate mechanisms, and none can be ruled out. The first is the antigen itself: chemical features of an antigen may favor one type of T cell activation over another, and there is mouse data suggesting certain compounds preferentially activate Th2 cells. The second is the antigen-presenting cell. Murine work had shown that different cell types — B cells, Kupffer cells, brain capillary endothelial cells — create different stimulatory environments, and Th1 and Th2 clones respond differently to them. It is plausible that antigens from Mycobacterium tuberculosis and Toxocara reach different presenting cells in different anatomical locations. The third is the local cytokine environment at the moment of priming: interferon-gamma appears to drive Th1 differentiation, interleukin four drives Th2, and interleukin ten from Th2 cells can inhibit Th1 cytokine synthesis. The paper treats these as open questions. What it does resolve is the outcome: whichever of those mechanisms is at work, they produce durable, clonally stable cytokine programs in human T cells. The functional consequences connect directly to clinical immunology. Interferon-gamma produced by Th1 cells drives delayed-type hypersensitivity — the skin response that a PPD test is designed to reveal. Interleukin four and interleukin five produced by Th2 cells drive IgE production and eosinophil recruitment, the cellular machinery of allergy and anti-parasite defense. The Th1 and Th2 split corresponds to recognizable immune phenotypes that clinicians already measure. Del Prete and colleagues showed those two tracks can coexist in the same individual, each shaped and stabilized by a different pathogen. Before this paper, the Th1 and Th2 framework in humans rested on indirect evidence — elevated cytokines in disease states, population-level patterns, and extrapolation from mouse experiments. This study provided something cleaner: one hundred twenty-nine clonally confirmed, antigen-specific human T cells, assessed individually under two distinct stimulation conditions, tracked over six months, with transcriptional data backing the functional readouts. The separation held at every level of analysis. What Del Prete and colleagues demonstrated is that the infectious history of a host can carve two enduring tracks into its immune repertoire — and that once a track is laid down, it does not waver. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.