CD44 and CD62L are two surface proteins that, when measured together, sort T cells into functionally distinct categories. A T cell that has never encountered its target looks different from one that fought off an infection last year, and both look different from one actively battling a pathogen right now. By checking whether CD44 is high or low and whether CD62L is present or absent, researchers can identify naive T cells, central memory T cells, and effector memory T cells in a single laboratory readout. This two-marker system has become one of the most widely used classification tools in immunology, but the molecules themselves do far more than serve as convenient labels.
How Two Markers Create a Map of T Cell Subsets
The standard approach uses antibodies against CD44 and CD62L in flow cytometry to separate T cells into three (sometimes four) populations based on which combination of these markers each cell carries. In mouse studies, the canonical breakdown looks like this:
- Naive T cells: CD44 low, CD62L high. These cells have never been activated by their specific antigen.
- Central memory (TCM): CD44 high, CD62L high. These cells were activated in the past, developed a memory of the encounter, and retained the ability to home back to lymph nodes.
- Effector memory (TEM): CD44 high, CD62L low. These are experienced cells that patrol peripheral tissues rather than recirculating through lymph nodes.
A standard flow cytometry protocol confirms this layout: anti-CD44 and anti-CD62L antibodies produce plots showing CD44⁻CD62L⁺ naive cells, CD44⁺CD62L⁺ central memory cells, and CD44⁺CD62L⁻ effector memory cells in both splenic and circulating T cell populations.1STAR Protocols. Protocol to analyze T cell activation, regulatory T cells, and cytokine production in mouse lymphoid organs using flow cytometry The logic is straightforward: CD44 goes up when a T cell gets activated and stays up permanently on memory cells, while CD62L gets cut from the surface when a T cell shifts toward an effector role. That combination of a marker that goes up and one that goes down creates four quadrants on a plot, three of which correspond to well-defined functional states.
What CD62L Does and Why Losing It Matters
CD62L, also called L-selectin, is a surface protein that acts as a homing beacon for lymph nodes. It belongs to a family of adhesion molecules that help immune cells grab onto the walls of specialized blood vessels and roll along them before squeezing through into lymph node tissue. The protein’s outermost domain interacts with sugar-containing molecules on the surface of those blood vessel walls, a process that is the first physical step a T cell takes when entering a lymph node.2PubMed Central. L-selectin: A Major Regulator of Leukocyte Adhesion, Migration and Signaling Without CD62L, a T cell cannot efficiently enter lymph nodes through this route.
This is why the presence or absence of CD62L carries so much information about a T cell’s job description. Naive T cells, which spend their time circulating between blood and lymph nodes waiting to encounter an antigen, maintain high CD62L. Central memory T cells, which also patrol lymph nodes watching for a familiar threat, keep their CD62L as well. But effector memory T cells, which need to be in peripheral tissues like the skin, gut, and lungs where infections actually happen, have shed their CD62L and largely stopped visiting lymph nodes.
The loss of CD62L is not a gradual fade. It is an active cleavage event. When a T cell recognizes its target antigen, enzymes on the cell surface physically cut CD62L from the membrane, releasing a soluble fragment into the surrounding fluid. This shedding has been directly linked to the T cell acquiring the ability to kill. In experiments with tumor-reactive human T cells, contact with matched tumor cells triggered rapid loss of CD62L, and this loss correlated with the appearance of a degranulation marker indicating the cell was now cytotoxic.3PubMed Central. The shedding of CD62L (L-selectin) regulates the acquisition of lytic activity in human tumor reactive T lymphocytes When researchers engineered T cells to carry a version of CD62L that resisted being cut, those cells showed impaired killing ability even after recognizing their target antigen. In other words, CD62L shedding is not just a byproduct of activation. It appears to be functionally connected to a T cell’s transition into a killer.
CD44 as More Than a Memory Label
CD44 is an adhesion molecule that binds hyaluronic acid, a major component of the tissue matrix surrounding cells throughout the body. After a naive T cell gets activated, CD44 levels on its surface increase and remain elevated permanently, which is why high CD44 expression has become the go-to marker for distinguishing experienced T cells from naive ones.4PubMed Central. CD44 regulates survival and memory development in Th1 cells But calling CD44 just a “memory marker” understates what the molecule actually contributes.
For starters, CD44 helps T cells physically get where they need to go. T cell migration into inflamed tissues depends on adhesion to hyaluronic acid, and CD44 is the main receptor that handles that interaction. Exposing T cells to inflammatory signals rapidly induces CD44 to bind hyaluronic acid, which allows the cells to stick to and move through inflamed tissue.5PubMed Central. Induction of interactions between CD44 and hyaluronic acid by a short exposure of human T cells to diverse pro-inflammatory mediators This gives effector and memory T cells with high CD44 levels a practical advantage: they are better equipped to migrate into sites of active infection or damage.
Beyond migration, CD44 contributes to T cell survival and function. Research has shown that engaging CD44 can boost T cell activation after antigen encounter and promote cell survival, and it also plays a role in regulating the contraction phase when an immune response winds down and most effector cells die off.6PubMed Central. Regulation of Antigen-Experienced T Cells: Lessons from the Quintessential Memory Marker CD44 Studies in other cell types suggest that the signaling pathways CD44 engages vary depending on context, which may explain how a single molecule can influence such different processes as activation, survival, and tolerance.7PubMed Central. Multifaceted regulation of T cells by CD44
Phenotypic Differences Between Naive and Memory T Cells Go Deeper Than Two Markers
CD44 and CD62L are the most commonly used pair, but they sit within a broader constellation of surface changes that distinguish naive from memory T cells. Compared to naive cells, memory T cells express higher levels of several integrins and adhesion molecules, including CD11a, CD29, CD49d, CD54, and CD58. Changes in CD45 isoforms also differ between the two populations: naive T cells display the highest molecular-weight version of CD45 (known as CD45RA in humans), while memory cells shift toward lower molecular-weight forms.8PubMed Central. Qualitative differences between naïve and memory T cells – Section: Phenotypic differences between naïve and memory T cells
What makes CD44 and CD62L especially useful is not that they capture every phenotypic difference, but that they are reliable, relatively binary, and easy to measure together. A naive T cell is consistently CD44 low and CD62L high. Central memory cells consistently add CD44 without dropping CD62L. Effector memory cells consistently show high CD44 and low CD62L. Other surface markers can vary more with tissue location, timing, or individual variability, which makes them less practical for a quick classification.
Tissue-Resident Memory Cells and the Limits of the Standard Framework
The three-subset model works well for T cells circulating in blood or present in lymphoid organs, but it misses an important population: tissue-resident memory T cells. These cells occupy tissues like the skin, lungs, and gut without recirculating through the bloodstream at all. They provide a rapid first line of defense when a previously encountered pathogen shows up again at a body surface, accelerating pathogen clearance before circulating memory cells even arrive.9PubMed Central. Tissue-resident memory T cells
Because tissue-resident memory cells do not recirculate into blood, they are absent from standard blood draws and are poorly represented in most flow cytometry panels. They are transcriptionally and phenotypically distinct from both central and effector memory T cells, so attempting to force them into the CD44/CD62L framework does not cleanly work. Researchers studying these cells rely on additional markers like CD69 and CD103 to identify them in tissue samples. This is a genuine gap in the two-marker classification system: it was designed for cells that move through blood and lymph, and tissue-resident cells simply do not play that game.
How These Markers Behave Differently in Mice and Humans
Most of the CD44/CD62L classification work comes from mouse studies, and the translation to human immunology is not seamless. In mice, CD44 is a clean divider: naive T cells express very little, and any cell with high CD44 has almost certainly been activated at some point. In humans, CD44 is more broadly expressed and less binary, which makes it a less reliable standalone marker for activation history. Human immunologists often lean on CD45RA and CD45RO isoforms, or CCR7, to make the same distinctions that CD44 handles in mice.
CD62L, by contrast, is a reasonably consistent marker across both species. Naive T cells in both mice and humans uniformly express high levels of CD62L and CCR7, and some memory cells in both species lose one or both of these lymph node-homing molecules.8PubMed Central. Qualitative differences between naïve and memory T cells – Section: Phenotypic differences between naïve and memory T cells The conceptual framework transfers well, but the specific antibody panels researchers use differ between species, and results from mouse experiments should not be assumed to look identical in human blood samples.
Aging Shifts the Balance Between Subsets
As organisms age, the distribution of T cell subsets defined by these markers changes substantially. The naive compartment shrinks, partly because the thymus (which produces new naive T cells) involutes with age, and partly because accumulated antigen exposures over a lifetime convert naive cells into memory populations. The result is a progressive shift toward CD44-high populations and, in many tissues, a decline in cells expressing high CD62L.
This shift has functional consequences. Work in aged tumor-bearing mice identified a specific CD44-low, CD62L-low CD8⁺ T cell subset that was generated inefficiently in older animals. In vitro experiments showed that stimulating naive CD8⁺ T cells first produces this subset before generating typical effector and memory populations, suggesting it represents an early transitional state. Cells in this subset expressed a distinctive set of genes related to one-carbon metabolism, which is important for antigen-specific T cell activation and mitochondrial function.10PubMed Central. Critical role of the CD44lowCD62Llow CD8+ T cell subset in restoring antitumor immunity in aged mice The poor generation of this subset in aged animals correlated with weaker antitumor immunity, pointing to a concrete way that age-related changes in the CD44/CD62L landscape can compromise immune function.
Aging also alters how T cell subsets respond to stimulation. Effector and memory CD8⁺ T cells that accumulate with age and carry the CD44-high phenotype show reduced responsiveness to certain signaling molecules compared to their naive, CD44-low counterparts.11The FASEB Journal. CD8(+), but not CD4(+) effector/memory T cells, express the CD44(high)CD45RB(high) phenotype with aging, which displays reduced expression levels of P2X(7) receptor and ATP-induced cellular responses So the aged immune system is not just short on naive cells; the memory cells it has accumulated may also function differently from their younger equivalents.
Cancer Immunotherapy and Why Subset Composition Matters
The CD44/CD62L framework has practical significance for cancer treatment, particularly for therapies that involve growing T cells outside the body and infusing them back into patients. The subset composition of the infused product affects how well the treatment works. Cells that retain CD62L and a central memory phenotype tend to persist longer after infusion and maintain the ability to self-renew, while cells that have fully differentiated into effector memory populations are potent killers but often exhaust quickly and disappear.
The connection between CD62L shedding and cytotoxic function creates a tension in this space. As described earlier, T cells that recognize tumor antigens shed CD62L as they acquire killing ability. But researchers found that engineering T cells to resist CD62L shedding impaired their cytotoxic activity, even though retaining CD62L would theoretically let the cells keep homing to lymph nodes.3PubMed Central. The shedding of CD62L (L-selectin) regulates the acquisition of lytic activity in human tumor reactive T lymphocytes This means you cannot simply lock CD62L in place to get the best of both worlds. The shedding event appears to be part of the functional program that arms the cell to kill.
Research on T cells in different body compartments adds nuance to this picture. Analysis of T cell phenotypes across organs found that effector and effector memory T cells, which are CD62L-low, were more prevalent in peripheral organs than in the spleen, and these CD62L-low populations showed higher expression of the exhaustion marker PD-1.12Journal for ImmunoTherapy of Cancer. Differential phenotypes of memory CD4 and CD8 T cells in the spleen and peripheral tissues following immunostimulatory therapy For adoptive cell therapy, this suggests that where the T cells end up in the body influences their functional trajectory, and the CD62L status they carry at infusion helps determine that destination.
CD44 Variants and Autoimmune Disease
CD44 is not a single uniform molecule. The gene encoding it can be spliced in different ways, producing variant forms that differ in their extracellular region. These variants, denoted by the specific exons they include (v1 through v10), are expressed differently across tissues and disease states. In the context of autoimmune disease, CD44 splice variants have drawn attention as potential players in pathological inflammation.
Work on multiple sclerosis provides a concrete example. Researchers found that T cells from the cerebrospinal fluid of people with MS expressed a specific splice variant, CD44v7, and that patients with high levels of this variant showed active inflammatory brain lesions significantly more often than those with low levels.13iScience. A TLR/CD44 axis regulates T cell trafficking in experimental and human multiple sclerosis This links alternative splicing of CD44 to the trafficking of T cells into the central nervous system during active disease, suggesting that the standard “CD44 high versus low” classification misses important biology encoded in which version of CD44 a cell expresses.
This finding points to a broader theme: CD44 and CD62L provide a useful first approximation, but the immune system’s complexity outgrows any two-marker system. As tools for measuring surface proteins become more sophisticated, with panels now routinely measuring 20 or more markers simultaneously, the original CD44/CD62L quadrant plot is being supplemented rather than replaced. It remains the backbone of T cell classification in countless experiments, but researchers increasingly layer additional markers on top to resolve finer distinctions that the two-marker system cannot capture on its own.