The
fucose post-translational modification is frequently increased in
pancreatic cancer, thus forming the basis for promising
biomarkers, but a subset of
pancreatic cancer patients does not elevate the known
fucose-containing
biomarkers. We hypothesized that such patients elevate
glycan motifs with
fucose in linkages and contexts different from the known
fucose-containing
biomarkers. We used a database of
glycan array data to identify the
lectins CCL2 to detect
glycan motifs with
fucose in a 3' linkage; CGL2 for motifs with
fucose in a 2' linkage; and RSL for
fucose in all linkages. We used several practical methods to test the
lectins and determine the optimal mode of detection, and we then tested whether the
lectins detected
glycans in
pancreatic cancer patients who did not elevate the
sialyl-Lewis A glycan, which is upregulated in ∼75% of pancreatic
adenocarcinomas. Patients who did not upregulate
sialyl-Lewis A, which contains
fucose in a 4' linkage, tended to upregulate
fucose in a 3' linkage, as detected by CCL2, but they did not upregulate total
fucose or
fucose in a 2' linkage. CCL2 binding was high in cancerous epithelia from pancreatic
tumors, including areas negative for
sialyl-Lewis A and a related motif containing 3'
fucose,
sialyl-Lewis X. Thus,
glycans containing 3'
fucose may
complement sialyl-Lewis A to contribute to improved detection of
pancreatic cancer. Furthermore, the use of panels of recombinant
lectins may uncover details about glycosylation that could be important for characterizing and detecting
cancer.