Granzyme K rises with lupus activity and maps to inflammatory immune circuits
Plasma granzyme K is emerging as more than a bystander in systemic lupus erythematosus. In this multi-omics study, circulating GZMK was higher in patients with SLE than in healthy controls and tracked with clinical disease activity, linking a cytotoxic lymphocyte protease to the broader inflammatory burden of the disease.
The signal did not stop in blood. Integrated single-cell analysis across tissues pointed to preferential GZMK enrichment in CD8+ effector memory T cells, with that compartment appearing more abundant in SLE. Protein-level and tissue-level analyses reinforced the same pattern, suggesting that GZMK marks a recurrent inflammatory T-cell program rather than a narrow, isolated laboratory finding.
A circulating protease that mirrored tissue inflammation
SLE is a disease built on immune overactivation, autoantibody formation, and organ injury, but the molecular links between circulating immune activity and tissue damage remain frustratingly incomplete. This study approached that gap from several angles at once: plasma testing, single-cell RNA sequencing, proteomics, multiplex immunofluorescence, animal models, and primary human B-cell experiments. The design is important because it allowed the investigators to trace GZMK from the bloodstream into skin, kidney, and immune-cell networks.
In the plasma cohort, 44 adults with SLE and 21 matched healthy controls were studied. The SLE group had a mean age of 38.80 years, 37 women and 7 men, and a median disease duration of 57 months. Half had lupus nephritis, 59.1% had rash, and 52.3% were anti-dsDNA positive. Mean SLEDAI-2K was 12.23, and 65.9% met the study definition of high disease activity. Those baseline features frame the central clinical question: can a single protease reflect both systemic activation and target-organ involvement?
The answer from this dataset leans yes. Circulating GZMK was associated with disease activity, and cell–cell communication analysis suggested enhanced predicted signaling between effector-memory T cells and B-lineage cells. In parallel, circulating GZMK was associated with a plasma cell-like population, tying the molecule to the antibody-producing arm of lupus immunopathology.
CD8 memory T cells and B-lineage dysfunction moved in tandem
The biology here is plausible. GZMK is not just a cytotoxic effector; it has recognized inflammatory functions outside classical target-cell killing. In inflamed tissue, GZMK-producing CD8 cells can behave like chronic immune agitators, sustaining cytokine-rich microenvironments and promoting downstream activation of neighboring cells. In lupus, where B cells are already prone to aberrant differentiation, that type of signal could amplify plasma-cell formation and feed the autoantibody loop.
The study’s single-cell and proteomic data fit that framework. GZMK was preferentially enriched in CD8+ Tem cells, while multiplex immunofluorescence provided evidence for a GZMK-enriched CD8 memory T-cell population in affected tissue. The same immune axis appeared to intersect with B-lineage cells. Instead of treating T-cell inflammation and B-cell dysregulation as separate problems, the findings connect them through a shared inflammatory program.
This matters at the bedside because lupus activity often looks different in the skin, kidney, blood, and serologic profile, yet the biology is often converging underneath. A marker that rises with both systemic activity and tissue injury offers a more integrated view of disease behavior.
Pharmacologic inhibition dampened lupus-like inflammation in mice
The mechanistic work moved beyond association. In two murine lupus models, PPACK, a compound that inhibits GZMK catalytic activity but is not selective for GZMK alone, reduced lupus-associated inflammatory phenotypes and partially normalized splenic B-cell composition. The compound’s lack of selectivity limits mechanistic certainty, but the direction of effect is hard to ignore: suppressing the enzymatic activity linked to GZMK was accompanied by improvement in immune abnormalities.
One sentence captures the translational weight of the experiment: interfering with this protease changed lupus-like disease features in vivo.
The mouse data also support the idea that GZMK sits upstream of B-cell skewing rather than merely reflecting it. Partial correction of splenic B-cell composition suggests that the protease may influence the balance of B-cell subsets that support autoantibody production and chronic inflammation. That is a clinically relevant target in a disease where excessive B-cell differentiation drives much of the damage.
Human B cells responded directly to active GZMK
Primary human B-cell experiments added another layer. Active recombinant GZMK augmented plasma-cell-associated phenotypic and transcriptional responses under B-cell expansion conditions. This finding closes the loop between association and function: GZMK was not only present near dysregulated B cells, it could actively push them toward a plasma-cell-like state.
The implications are straightforward. If GZMK contributes to plasma-cell differentiation or survival, it could help sustain the autoantibody machinery that defines lupus immunobiology. That would also make the protease relevant to serologic activity, complement consumption, and organ-specific inflammation, all of which are central to clinical monitoring.
The study’s own framing places GZMK at the intersection of inflammatory T-cell programs, complement-linked injury, and B-cell/plasma-cell abnormalities. Prior work has suggested that GZMK can act as an extracellular pro-inflammatory protease and may even influence complement pathways. In a disease defined by immune-complex deposition and complement activation, that overlap is biologically attractive, even if the current study does not prove every step in the causal chain.
What this means for lupus biomarker work
For clinicians, the immediate relevance is biomarker logic rather than treatment change. GZMK is not ready for routine testing on the basis of a single-center cross-sectional study, and the animal experiments used a nonselective inhibitor. Still, the data support a more integrated view of lupus immunopathology: a circulating protease, enriched in CD8 memory T cells, was linked to disease activity, tissue inflammation, and B-cell dysregulation across multiple platforms.
The cohort size was modest, and the study was cross-sectional, so it cannot establish temporal ordering or tell whether GZMK rises before flares, falls with treatment, or predicts organ-specific outcomes. Skin and kidney analyses were based on small tissue sets, and the inhibitor used in mice cannot isolate GZMK alone. Even so, the consistency across plasma, single-cell, proteomic, tissue, and functional assays gives the signal unusual coherence.
The next question is not whether lupus contains inflammatory T-cell and B-cell circuits. It clearly does. The sharper question is whether GZMK can be developed into a usable biomarker of activity or a tractable point of intervention in patients with cutaneous, renal, or systemic disease.