Breast cancer, a disease where malignant cells form in breast tissue, is the most common cancer in women in the U.S. For 2022, the National Breast Cancer Foundation estimates that there will be 338,900 new cases of invasive and non-invasive breast cancer diagnosed in the U.S. alone. There are dozens of types of breast cancer where treatment options can differ depending on type. While genetics is thought to be a major role in the development of breast cancer in 5–10% of all cases, there are many other factors that can be examined in the context of breast cancer. One of these factors is protein glycosylation.
Half of all human proteins are glycosylated. In cancer, changes to glycan composition of proteins are correlated with tumor proliferation, metastasis, and immune evasion. Thus, it is not surprising that the majority of FDA-approved cancer biomarkers are glycoproteins or carbohydrate antigens.
The study of glycosylation in breast cancer began in the early 1990s when scientists identified a correlation with altered glycosylation (1). Since then, scientists have adopted many techniques to further our understanding of glycosylation in cancer and work towards therapeutic outcomes. Check out our related blog post to learn more about the impact of glycosylation in cancer research.
Sialylation is the addition of sialic acid to glycoproteins. In many cancers, elevated levels of sialylation promote tumor progression. One type of sialylation called polysialylation generates long glycan chains composed of α2,8-linked sialic acid residues known as polysialic acid or polySia.
In healthy adults, polySia expression is limited to specific cell types and protein carriers such as neurons and leukocytes (1). However, neoexpression of polySia occurs in neuroblastoma, glioma, lung carcinoma, and leukemia (1). Overexpression of polySia is thought to increase cancer cell migration and invasion. PolySia was identified by a National Cancer Institute pilot project as a high-priority cancer antigen for study in 2009 (2).
Below, we highlight a recent work from the University of British Columbia on the role of polySia, CD56, and polysialyltransferases in the breast tumor microenvironment (3). More specifically, they examined the expression of polySia, CD56, and polysialyltransferases in tumor cells and in tumor infiltrating lymphocytes, making use of immunocytochemistry reagents from Vector Laboratories including VECTASTAIN® ABC-AP Kit, Goat anti-Rabbit IgG Antibody (H + L), Peroxidase, Horse anti-Mouse IgG Antibody (H + L), Peroxidase, DAB (3-3’-Diaminobenzidine) Substrate, and VectaMount® Mounting Medium. The findings from their expression studies were paired with patient data to address one question: Do these molecules hold prognostic potential?
PolySia expression studies
The authors first examined how polySia expression differed in primary breast tumors versus normal and adjacent normal breast tissue. Using immunohistochemistry and tissue microarrays, they found that polySia is expressed in breast tumor cells at much higher levels than normal tissue. This was true for all four breast cancer subtypes, and there were no significant differences in polySia levels between the subtypes.
They also found polySia expression on tumor infiltrating lymphocytes (TILs). This was the first time that polysialylated TILs were identified in the tumor microenvironment.
The researchers found that polySia levels in tumor cells could not predict survival outcomes for individuals with low or moderate polySia expression compared to those with high polySia expression. There was no correlation with patient outcomes over a 10-year period though high polySia levels were associated with a worse outcome within a 5-year period.
In TILs, the researchers saw that high levels of polysialylated TILs were correlated with poor survival. Compared to tumors that were TIL-negative, individuals with low or moderate polysialylated TIL had increased overall survival.
Polysialyltransferase expression studies
Two sialyltransferases, ST8Sia2 and ST8Sia4, synthesize polySia. To evaluate their expression in the tumor microenvironment, the scientists used in situ hybridization probing for each sialyltransferase. They found that in tumor cells and infiltrating stromal cells, ST8Sia4 was the main polysialyltransferase expressed and its expression was not associated with a specific breast cancer subtype. However, ST8SIa4 expression in tumor cells and infiltrating stromal cells was correlated with a higher tumor grade.
While the researchers saw an increase in the sialyltransferase expression, they found no correlation between ST8Sia4 expression and polySia expression.
The researchers found that high ST8Sia4 expression in tumor cells was associated with poor overall survival.
In tumor-infiltrating lymphocytes, low levels of ST8Sia4 expression were correlated with worse patient outcomes. High levels of ST8Sia4 positive-infiltrating stromal cells were predictive of a good 10-year prognosis. Yet surprisingly, the authors found that high levels of polysialylated TILs were associated with poor outcomes. Why this discrepancy between ST8Sia4 and polysialylation in TILs? It’s possible that there are different lymphocyte subpopulations where polysialylated proteins have various prognostic indications or it could be because ST8Sia4 RNA levels are not representative of protein levels.
CD56, a phenotypic marker of natural killer cells, is a carrier protein for polySia chains. However, CD56 and polySia dual immunofluorescence staining (with the help of the Vector® TrueVIEW® Autofluorescence Quenching Kit) found that not all polySia expressing tumors were CD56 positive. While 60% of tumors were positive for polySia, only 15% of tumor cells express CD56. For CD56 positive tumors, only 55% were also polySia positive. These results raise the possibility that other proteins are polysialylated in breast cancer.
Although CD56 did not correlate with polySia expression, CD56 did correlate with HER2 status.
CD56 expression correlated with both HER2+ tumors and luminal B tumors and its level was associated with poor patient outcomes.
In HER2 expressing tumors, high levels of CD56 correlated with poor patient outcomes. CD56 and polySia expressing TILs in patients were a good prognostic indicator while CD56 negative and polySia positive TILs were a good prognostic indicator.
The investigation of polysialylation in breast cancer adds to a wealth of studies that illustrate the complex role glycosylation has in cancer. In the case of polySia, different polysialylated proteins on different cell types appear to be indicative of different patient outcomes. Polysialylation in some cells seems to have negative effects on prognosis while polysialylation in other cells appears to have positive impacts. It’s also possible that polysialylation affects CD56-mediated tumor cell killing, but this has so far been uncharacterized. As this study was the first comprehensive analysis of polySia, CD56, and polysialyltransferase in breast cancer, many questions remain for exploration, including whether these expression patterns can be used to predict patient prognosis.
Be sure to check out our Glycobiology Resources page for tools to help you unlock deeper insights in your research, and stay tuned for the latest updates from the blog.