Mutant calreticulin is dependent on zinc to drive constitutive MPL activity

Mutations of the calreticulin gene (CALR) have been recognized as driver mutations involved in the development of myeloproliferative neoplasms (MPN) along with Janus kinase (JAK) or thrombopoietin receptor (MPL) mutations. CALR mutations in MPN are insertion-deletion mutations, and the most common is a 52-bp deletion (CALR^del52).

CALR^del52 homomultimers are responsible for binding to MPL and causing constitutive activation of JAK-STAT signaling. However, there is limited understanding of the mechanism of CALR^del52 interaction with MPL and its activation. 

Rivera et al. recently conducted a comprehensive mutagenesis analysis and explored the mechanism of CALR^del52 binding and activation of the MPL signal, including the role of the glycan-binding lectin motif, which associates with the extracellular domain of MPL and the zinc-binding domain.¹ The results are published in Blood Advances, and the MPN Hub is pleased to summarize the key findings here.

Materials

293T cells were derived from the American Type Culture Collection, and MPL-expressing Ba/F3 cells were generated by retroviral transduction. Peripheral blood mononuclear cells were isolated from patients with MPN.

Results

Residues related with lectin and zinc-binding activity are essential for oncogenic activity based on a mutagenesis screen described below:

• As expected, nine residues involved in the function of the lectin motif abrogated cytokine independence. Specifically C105A, K111A, G133A, or D135A mutations displayed complete nullification of cytokine independence, decreased STAT3/5 phosphorylation, and prevented co-immunoprecipitation of CALR^del52 with MPL.
• Interestingly, tyrosine mutations Y109A and Y128A which inhibit normal CALR function, showed only partial loss of cytokine independence in CALR^del52 transfected cells but had no effect on STAT3/5 phosphorylation or MPL binding.
• When investigating the role of zinc-binding histidines, it could be shown that single mutations in H99A, H145A, and H170A (1His) showed partial abrogation of cytokine independence, while binding to MPL was maintained.
• However, loss of either two or all three histidine residues in combination (2His and 3His) led to significant impairment of cytokine-independent growth along with decreased pSTAT3/5 phosphorylation and loss of MPL binding.
• By contrast, mutating a histidine not involved in zinc-binding, H42A, did not abrogate cytokine-independent growth and the capacity to promote STAT3/5 phosphorylation.
• Using mCherry-tagged CALR^del52 and green fluorescent protein (GFP)-MPL, significant co-localization could be shown for MPL with CALR^del52 and with 1His-CALR^del52 variants, but not with wild-type CALR or 2His and 3His CALR^del52 variants.

Histidine-mutated CALR^del52 is incapable of homomultimerization:

• FLAG-tagged CALR^del52 co-immunoprecipitated with a V5-tagged CALR^del52 but did not bind to wild-type CALR.
• Multimerization was retained in lectin-deficient CALR^del52 and all three 1His- CALR^del52 variants. However, it was impaired in 2His- and 3His- CALR^del52 variants.
• Lectin-deficient CALR^del52-D135A retained the ability to homomultimerize with another CALR^del52-D135A, but not with histidine-mutated 3His-CALR^del52.

 Zinc is essential for CALR^del52-MPL complex formation and downstream signaling:

• Using affinity chromatography with a zinc resin, the authors found retention of CALR^del52 and 1His- CALR^del52 in the zinc-bound fraction, which was not the case for wild-type CALR, 2His-, and 3His- CALR^del52, suggesting that zinc may be essential for CALR^del52 multimerization.
• This was further tested in co-precipitation experiments using 293T cells expressing FLAG- and V5-tagged CALR^del52 and MPL. Treatment with Tetrakis (2-pyridylmethyl)-ethylenediamine (TPEN), a zinc chelator, decreased CALR^del52 multimerization and impaired MPL binding, confirming that zinc was needed for CALR^del52 to multimerize and interact with MPL.
• To understand if zinc chelation with TPEN could eradicate cytokine independence and JAK-STAT signalling, CALR^del52-expressing Ba/F3-MPL cells were investigated. Treatment with TPEN led to decreased viability and diminished STAT3/5 phosphorylation. Similar effects were seen using another zinc chelator, clioquinol.
• Furthermore, there was reduced interaction between CALR^del52 and MPL in these cells as seen in FLAG pulldown experiments.
• By contrast, Ba/F3 cells transformed with the oncogenic MPL^W515L were less sensitive to TPEN-induced cytotoxicity and had no impairment in STAT3/5 phosphorylation.

TPEN preferentially eradicates CALR-mutant primary MPN cells:

• No evidence of impaired growth of wild-type colonies were observed at TPEN dosages <3 µM.
• When testing peripheral blood mononuclear cells (PBMCs) from five patients with CALR^del52-mutated myelofibrosis, there was clear indication that zinc chelation selectively affected the growth of CALR^del52-positive, patient-derived, burst-forming unit erythroid (BFU-E) colonies. Using supplementation with 2.5 µM TPEN, the authors demonstrated a decreased growth of CALR^del52-positive BFU-Es, but a normal growth of wild-type BFU-Es in four of five patients.

Conclusion

This study provides a better understanding of the molecular mechanism through which CALR^del52 exerts its oncogenic activity. The study identifies the zinc-binding histidines within CALR^del52 essential for the formation of homomultimers that interact with MPL to induce its constitutive activation. It also shows that homomultimerization and the MPL-binding of CALR^del52 are zinc-dependent and could therefore, be a potential target for treatment of CALR-mutant MPNs.