The growth of human colon cancer cells (raised in culture) is unaffected by the MELK gene. Top row: untreated cells; bottom row: cells treated with a MELK inhibitor drug. Left two columns, control cells; right two columns, cells in which MELK gene has been knocked out. CSHL researchers conclude that MELK is not involved in cancer proliferation. Photo: Sheltzer Lab, CSHL

About a decade ago, dozens of scientific teams identified a protein kinase called the Maternal Leucine Zipper Kinase, or MELK, as a culprit in cancer – and a potentially revolutionary drug target. Drug investigation commenced, and clinical trials are still underway targeting the pathway.

But the scientific tide has now turned, most recently with a paper in eLife by scientists from Cold Spring Harbor Laboratory.

Their gene-editing techniques are “self correcting” the science behind MELK, they explain.

“We thought we would eliminate MELK and show that it killed cancer cells. Then we would know our CRISPR techniques were working,” said Jason Sheltzer, a fellow at Cold Spring Harbor Laboratory, one of the authors. “But, to our great surprise, the cancer cells didn’t die. They just didn’t care.”

The team, including personnel from Stony Brook University and Google, have taken several years to perform genomic analyses on extracted tumor tissue.

Their targets were genes that showed up most, in the worst outcomes. The genes associated with the lowest survival rates were identified. MELK was especially very highly expressed in patients who generally died shortly after the tumors were taken.

The next step was to target those through the gene editing process CRISPR. If MELK was removed, they figured, the cell lines would stop growing.

But instead, in both in vitro and in vivo mice models, the cancer was unperturbed – and continued its progression.

The experiments attempted to isolate the cellular function of MELK based on stem cell renewal, oxygen sensing, chemotherapy resistance, and other factors, the researchers report. But they found that the early findings about the protein kinase could have simply been attributed to MELK, while it was actually “off target effects” like RNA interference that may have been hitting other unintended genes.

“Although MELK has been implicated in each of these processes, our results demonstrate that MELK-knockout cancer cell lines grow at wild-type levels in a variety of assays designed to test these pathways,” the study authors write. “We speculate that, as has previously been reported for one MELK inhibitor and one set of MELK-targeting shRNA’s, several previous studies of MELK function may have been compromised by off-target activity of the constructs and inhibitors that were used.

“We believe that, if MELK does play a role in cancer, it may be detectable only in very limited circumstances, and likely in vivo,” they add.

The same team had demonstrated in a paper last year that triple-negative breast cancer continued to grow in vitro despite the knockout of MELK.