We will here only focus on major issues and overlook minor errors.
Evidently, Linder and Shoshan do not believe that the cyclolignans picropodophyllin (PPP) and podophyllotoxin (PPT) are inhibitors of the insulin-like growth factor-I receptor (IGF-IR), reported by us 3 years ago (1). We also reported that they are selective in the sense that they do not coinhibit the highly homologous insulin receptor. This is crucial because coinhibition has hampered the development of IGF-IR inhibitors for anticancer therapy. PPP also showed selectivity in inducing antitumor activity in IGF-IR–positive xenografts (1, 2) but not in IGF-IR–negative ones (1). Whereas PPT is also an established microtubule inhibitor (3), PPP has previously been considered to almost lack biological activity (3). A recent in vivo study shows that daily treatment with large doses of PPP is well tolerated and suppresses an aggressive systemic malignancy for several months.1
Vanderkerken, Jernberg-Wiklund et al., in preparation.
Linder and Shoshan support their statements by repeating one of our experiments using IGF-IR–deficient mouse embryonic fibroblasts (R−). Growth inhibition (∼50%) was induced by 0.5 μmol/L PPP and PPT. These results differ from ours (1). We could not obtain 50% growth inhibition (IC50) for the two compounds even at 15 μmol/L. Linder and Shoshan conclude that PPP and PPT do not induce cell death via IGF-IR.
However, established mouse cell lines are changeable and often unpredictable over time. We and others have noted that the properties of R− have changed. These cells can now undergo spontaneous transformation. When we recently studied this in more detail, we discovered two different subclones of R− (4). One subclone (R−S), from later passages, surprisingly expresses a truncated form of IGF-IR β-subunit being constitutively phosphorylated, whereas the other (R−R) does not. Growth of R−S (probably used by Linder and Shoshan) is inhibited by small interfering RNA targeting IGF-IR and PPP. R−R cells (original R−, received from Dr. Baserga in 1996, and used by us in 2001–2004) are resistant to IGF-IR small interfering RNA and PPP. Notably, R− are rarely used to show specificity of new IGF-IR inhibitors, a fact that can be explained by our observation that these cells may express IGF-IR activity.
Regarding the cell cycle aspect, it is known that IGF-IR is also important for G2-M progression (5).
Linder and Shoshan believe that PPP induces cell death via microtubule inhibition and refer to older studies. However, PPP is frequently contaminated with 1% to 2% of PPT unless extensively purified and carefully handled (4). Thus, a small contamination with PPT may be the reason for previously observing a marginal effect of PPP on microtubule assembly.
We use ultrapure PPP (99.7%) and, recently, we showed that PPP (up to 50 μmol/L) does not bind to tubulin, whereas PPT does (4). Thus, it is unlikely that PPP induces cell death via microtubule inhibition.
In conclusion, we feel that this reply provides further support for our finding that cyclolignans can act as IGF-IR inhibitors.