The anticipated outcome on cell proliferation following downregulation of APLP2 and/or APP was used as the functional readout (fourth and fifth columns, Table I). Three basic questions were considered for the construction of the table. First, are APLP2 and APP in the same pathway? If APLP2 selleck chemical and APP are not in the same pathway, parallel pathways and unrelated pathways should be considered. As shown in Fig. 2, simultaneous downregulation of both APLP2 and APP did not enhance growth inhibition beyond down-regulation of APLP2 or APP alone, signifying that APLP2 and APP likely act upon the same pathway in pancreatic cancer cell growth (eliminating the scenarios in the bottom half of Table I).

Second, what is the relationship between APLP2 and APP within the pathway? In the same pathway, APLP2 and APP could act at the same point, or sequentially to one another, or in a parallel (or switchboard) relationship. The last question considered was whether APP and APLP2 have the same role. If APLP2 and APP cannot exchange for one another, they evidently have unique roles. Because growth inhibition was still observed when either APLP2 or APP was downregulated, APLP2 expression does not compensate for loss of APP in pancreatic cancer cell growth, and vice versa. Therefore, pathway relationships outlined in rows 1 and 4 from Table I do not match the experimental data (Fig. 2). Consequently, two models were devised (rows 2 and 3) that match the experimental data from Fig. 2. Table I Phenotypic outcomes of postulated pathways including APP and/or APLP2.

The second row of Table I indicates that APLP2 and APP could be involved at the same point within the same pathway, but yet serve unique roles; an example of this scenario would be a functional heterodimer of APLP2 and APP. Heterodimers of APLP2 and APP are known to form (57). In this context, a homodimer of APLP2 or APP could not recapitulate the function of the APLP2-APP heterodimer, and equal growth inhibition would occur with loss of one or both proteins. The scenario in the third row also matches our experimental data. Instead of APLP2 or APP acting at the same point in a pathway, one protein would be upstream of the other, and both would be required for a growth signal to be conducted. In both devised scenarios matching the experimental data, APLP2 and APP would have unique roles in the growth of pancreatic cancer cells.

Ultimately, GSK-3 the conclusion best fitting the data is that APLP2 and APP serve unique roles in the growth of pancreatic cancer cell lines. Downregulation of APLP2 C-terminal fragments (by siRNA or ��-secretase inhibitors) resulted in growth inhibition of S2-013 cells, despite maintained expression of APP full-length protein (Fig. 2A and B and data not shown); this is noteworthy since ��-secretase inhibitors are also capable of reducing cleavage of APP.