sign in sign up
<<
Home , Sorafenib

resistance and microcytic erythrocytosis in PNAS PLUS Δ a KitV558 ;T669I/+ gatekeeper-mutant mouse model of gastrointestinal stromal tumor

Benedikt Bosbacha, Shayu Deshpandea, Ferdinand Rossia, Jae-Hung Shiehb, Gunhild Sommerc, Elisa de Stanchinad, Darren R. Veachd, Joseph M. Scandurae, Katia Manova-Todorovaa, Malcolm A. S. Mooreb, Cristina R. Antonescuf, and Peter Besmera,1

aDevelopmental Biology, bCell Biology, and dMolecular Pharmacology and Chemistry Programs, Sloan-Kettering Institute, New York, NY 10065; fDepartment of Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY 10065; cDepartment of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, SC 29425; and eDepartment of Medicine, Weill Cornell Medical College, New York, NY 10065

Edited by Peter K. Vogt, The Scripps Research Institute, La Jolla, CA, and approved May 4, 2012 (received for review September 20, 2011) Most gastrointestinal stromal tumors (GISTs) harbor a gain-of- resistant to inhibition by imatinib. Imatinib binds to the inactive function mutation in the Kit receptor. GIST patients treated with conformation of the ABL and KIT and not to the active the tyrosine inhibitor imatinib frequently develop imatinib conformation and inhibits juxtamembrane domain KIT mutants resistance as a result of second-site Kit mutations. To investigate but not activation-loop KIT mutants (7). the consequences of second-site Kit mutations on GIST development That oncogenic KIT mutations have a critical role in the de- and imatinib sensitivity, we engineered a mouse model carrying in velopment of human neoplasias was strengthened by the obser- Δ the endogenous Kit locus both the KitV558 mutation found in a vation of familial GIST and familial mastocytosis (8). Patients familial case of GIST and the KitT669I (human KITT670I) “gatekeeper” with familial GIST also may have cutaneous mastocytosis and mutation found in imatinib-resistant GIST patients. Similar to hyperpigmentation. The observation of germ-line KIT gain-of- Δ Δ KitV558 /+ mice, KitV558 ;T669I/+ mice developed gastric and colonic function mutations provided us with a rationale for developing interstitial cell of Cajal hyperplasia as well as cecal GIST. In contrast a mouse model for familial GIST. The KIT-V558 deletion mu- V558Δ/+ to the single-mutant Kit control mice, treatment of the tation found in the first familial GIST case was introduced into MEDICAL SCIENCES V558Δ;T669I/+ Kit mice with either imatinib or failed to in- the mouse genome using a knockin strategy (9). The mutant hibit oncogenic Kit signaling and GIST growth. However, this re- animals developed ICC hyperplasia and neoplastic lesions in the V558Δ;T669I/+ sistance could be overcome by treatment of Kit mice cecum indistinguishable from human GIST with complete pen- with or sorafenib. Although tumor lesions were smaller etrance (10, 11). V558Δ;T669I/+ in Kit mice than in single-mutant mice, both interstitial Long-term imatinib treatment of GIST and of patients who cell of Cajal hyperplasia and mast cell hyperplasia were exacerbated have CML is associated with the development of drug resistance. V558Δ;T669I/+ V558Δ;T669I/+ in Kit mice. Strikingly, the Kit mice developed In GIST most cases of resistance appear to derive from second- a pronounced polycythemia vera-like erythrocytosis in conjunction site mutations in the kinase domain of the KIT receptor (12, 13). with microcytosis. This mouse model should be useful for pre- In patients who have CML, second-site mutations in BCR-ABL clinical studies of drug candidates designed to overcome imatinib are the predominant mechanism of drug resistance (14, 15). The resistance in GIST and to investigate the consequences of onco- second-site mutations in acquired imatinib-resistant GIST tend genic KIT signaling in hematopoietic as well as other cell lineages. to be single amino acid substitutions in KIT, located on the allele with the primary mutation (12, 16). Second-site mutations in soft tissue sarcoma | hematopoiesis | erythropoiesis | drug resistance GIST occur in catalytic domain II of KIT, exons 17 and 18, as well as in the N-terminal kinase domain, exon 13 (V654A) and astrointestinal stromal tumor (GIST) is the most common exon 14 (T670I) (12). In the gatekeeper T670I mutation, the Gmesenchymal tumor of the . GISTs isoleucine methyl group protrudes into the imatinib binding site express receptor KIT and are thought to derive and disrupts an important hydrogen bond formation between + low from a KIT or KIT interstitial cell of Cajal (ICC) progenitor imatinib and the kinase, precluding proper binding of imatinib or from ICCs themselves (1). The principal genetic events re- (7). Second-site mutations in the activation loop within the kinase sponsible for the pathogenesis of GIST are thought to be gain-of- domain stabilize the active conformation of KIT and maintain it function mutations in the KIT gene or in a small subset in the constitutively activated at a high level, thereby preventing imatinib PDGFR-alpha gene (2, 3). KIT-activating mutations in GIST are binding. Currently, several drugs, including sunitinib, dasatinib, found predominantly in the juxtamembrane domain of the KIT and sorafenib, are being evaluated for efficacy in the treatment receptor (exon 11) (4), but mutations in the extracellular (exon of imatinib-resistant GIST. Previous in vitro studies indicated 9) and kinase domains of KIT have been described as well (5, 6). The KIT juxtamembrane domain has an autoinhibitory role and stabilizes an inactive conformation of the KIT kinase; mutation of this domain disrupts the conformational integrity and thus KIT Author contributions: B.B., S.D., J.-H.S., G.S., E.d.S., M.A.S.M., C.R.A., and P.B. designed diminishes autoinhibition (7). activation-loop mutations found research; B.B., S.D., F.R., J.-H.S., G.S., J.M.S., and C.R.A. performed research; D.R.V. con- in acute myeloid , mast cell neoplasms, and seminomas tributed new reagents/analytic tools; B.B., S.D., F.R., J.-H.S., G.S., D.R.V., J.M.S., K.M.-T., stabilize an active conformation of the KIT kinase. M.A.S.M., C.R.A., and P.B. analyzed data; and B.B., S.D., J.M.S., M.A.S.M., C.R.A., and P.B. Imatinib mesylate, an inhibitor of the KIT, PDGFR, and wrote the paper. BCR-ABL tyrosine kinases, is the first-line therapy in patients Conflict of interest statement: P.B. received funding from Novartis for research on GIST. with chronic myelogenous (CML) and metastatic This article is a PNAS Direct Submission. GIST. Imatinib is most effective in GISTs with KIT-activating Freely available online through the PNAS open access option. mutations in the juxtamembrane domain, some kinase domain 1To whom correspondence should be addressed. E-mail: [email protected]. mutations, or extracellular domain mutations. However, KIT This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. mutations that destabilize the inactive form of the kinase are 1073/pnas.1115240109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1115240109 PNAS Early Edition | 1of8 Downloaded by guest on September 26, 2021 Δ that both sunitinib and sorafenib inhibit the T670I gatekeeper mice, double-mutant KitV558 ;T669I/+ mice had a prolonged lifespan mutation, but imatinib, dasatinib, and failed to do with a median survival of 14 mo (n > 43 each, P < 0.0001) so (17, 18). (Fig. 1B). Δ Because of the clinical importance of imatinib resistance, the Invariably, KitV558 ;T669I/+ mice developed cecal tumors. These Δ development of new strategies for the treatment of GIST is highly tumors were smaller than in KitV558 /+ mice, perhaps explaining relevant. Such strategies may be based on the development of the improved survival by a decreased chance of intestinal ob- KIT kinase inhibitors that show efficacy with the resistant forms struction (Fig. 1C). The average tumor diameter in 3-mo-old Δ Δ of KIT; targeting of downstream signaling components critical animals was fivefold smaller in KitV558 ;T669I/+ than in KitV558 /+ for oncogenic KIT function could provide a second approach. mice (1.4 ± 0.1 mm vs. 7.0 ± 0.3 mm, P < 0.001). Interestingly, However, models to examine these approaches and their possi- not only were the cecal tumors smaller, but the length of the Δ ble side effects in vivo have not been reported. Here we describe cecum was significantly shorter in KitV558 ;T669I/+ mice compared the derivation of a mouse model for imatinib-resistant GIST that V558Δ/+ ± ± V558Δ T669I with Kit and wild-type mice (13 2 mm vs. 24 2 mm, includes both the juxtamembrane domain Kit and Kit P = 0.003) (Fig. 1C). Histological analysis of the shortened Δ gatekeeper mutations as a tool to develop therapeutic strategies cecum of the KitV558 ;T669I/+ mice revealed an intact lumen and Δ for imatinib-resistant GIST and to investigate the consequences mucosa similar to those in KitV558 /+ mice. In contrast to the Δ of KIT oncogenic signaling in other KIT-dependent cell lineages overall reduction in tumor size in the cecum, KitV558 ;T669I/+ mice in particular in hematopoiesis. developed more pronounced ICC hyperplasia in the – V558Δ/+ Results (Fig. 2 C and I; also see Fig. S3Aa d) and colon than Kit Δ mice (Fig. 2 B and H; also see Fig. S3Ba–d). Immunohisto- Derivation and Phenotypic Characterization of KitV558 ;T669I/+ Gate- chemical analysis of gastric and colonic sections revealed KIT keeper Mice. To investigate the consequences of second-site staining as well as phosphorylation of S6, MAPK, and STAT3 KIT mutations on imatinib susceptibility and GIST development V558Δ/+ V558Δ;T669I/+ in vivo, we generated a mouse model introducing both the in the ICC hyperplasia of Kit as well as Kit Δ – – fi KitV558 and the KitT669I gatekeeper mutation, corresponding to mice (see Fig. S3 Ae n and Be n), but no signi cant differences T670I in signal transduction were apparent in the two strains that could human KIT and found in cases of imatinib-resistant GIST, Δ explain the exacerbated ICC hyperplasia in KitV558 ;T669I/+ mice. into the endogenous Kit locus. To facilitate simultaneous in- Δ H&E-stained sections of cecal tumor lesions in both KitV558 /+ troduction of the two point mutations into the mouse Kit gene, V558Δ;T669I/+ the targeting vector included a floxed neomycin-resistance gene and Kit mice showed a histology indistinguishable (NEO) cassette in Kit intron 11 for positive selection of from human GIST (Fig. 2 E and F). Δ recombinant ES cells containing both the V558 (exon 11) and V558Δ;T669I/+ T669I (exon 14) mutations (Fig. 1A). After successful integration Kit Gatekeeper Mutation Confers Resistance to Imatinib V558Δ;T669I-NEO and Dasatinib in Vivo. To investigate the in vivo sensitivity of the and germ-line transmission of the Kit allele, the V558Δ;T669I intronic NEO cassette was removed by crossing to Tg(EIIa-cre) Kit gatekeeper mutation to tyrosine kinase inhibitors Δ mice (19). The resulting KitV558 ;T669I allele retains a single loxP (TKIs) with KIT inhibitory potential, we treated cohorts of 3- to Δ V558Δ;T669I/+ site in intron 11 (Fig. 1A3). A KitV558 allele with a loxP site in 4-mo-old Kit mice with imatinib, dasatinib, suniti- Δ intron 11 was generated as a control for the KitV558 ;T669I/+ allele nib, or sorafenib. First, we analyzed signaling cascades known to (Fig. S1A). be affected by imatinib after short-term (6-h) drug treatments (10, Δ Double-mutant KitV558 ;T669I/+ mice are viable and fertile 11). Second, proliferation, , and histological changes Δ but, in contrast to KitV558 /+ mice, were born at sub-Mendelian within tumors and adjacent mucosa were assessed after long- ratios when crossed to wild-type mice (35% instead of 50% het- term (7-d) treatment with Kit inhibitors. In addition, changes in Δ erozygous offspring). In comparison with single-mutant KitV558 /+ the phosphorylation status of proteins in GIST following long-

A V558Δ T669I B C 100 Targeting vector:

+/+ +/+ DTA NEO 14 17 75 Kit genomic locus: V558Δ;T669I/+ 1) Wild-type (+) 50 exon 8 11 14 17 Survival [%] Survival 558Δ;669I/+ 2) V558Δ;T669I- 25 NEO NEO V558Δ/+ x Tg(EIIa-cre) 0

3) V558Δ;T669I 0 200 400 600 800 558Δ/+ 11 14 Age [days]

Fig. 1. Derivation and phenotypic characterization of Kit-gatekeeper mice. (A) Targeting strategy for the simultaneous knock-in of V558Δ (exon 11) and T669I (exon 14) into the 129/Sv Kit locus. Blue and red bars denote the exons with the respective point mutations. A similar targeting vector with a shorter 3′ homology arm was used to generate the single-mutant KitV558Δ/+ mice (Fig. S1A). Triangles (not drawn to scale) indicate loxP sites; white gaps indicate BamHI restriction sites; bar indicates 518-bp Southern blot probe. DTA, diphtheria toxin A gene; NEO, neomycin resistance gene. (B) Kaplan–Meier survival Δ Δ plot showing increased survival of gatekeeper-mutant KitV558 ;T669I/+ mice in comparison with KitV558 /+ mice (n ≥ 43 each; ticks indicate censored subjects). Δ (C) Photographs of ileocecal junctions showing reduced length and diameter of tumor alongside the cecum in KitV558 ;T669I/+ mice (Middle; red bracket Δ indicates straight cecal GIST) in comparison with KitV558 /+ mice (Bottom; blue bracket indicates twisted cecal GIST). Of note, the cecum is significantly shorter in KitV558Δ;T669I/+ mice than in wild-type mice (Top; black arrow) and KitV558Δ/+ mice. Representative pictures of 3-mo-old animals are shown with colon facing down left and ileum facing down right (n ≥ 59 each). (Scale bar, 1 cm.)

2of8 | www.pnas.org/cgi/doi/10.1073/pnas.1115240109 Bosbach et al. Downloaded by guest on September 26, 2021 PNAS PLUS +/+ V558Δ/+ V558Δ;T669I/+ the coexpression of one mutant and one wild-type KIT allele), as detected in the GISTs of most patients, results in the formation of functional KIT heterodimers driving oncogenic signaling. Assuming that imatinib inhibits wild-type KIT and possibly other Δ A BC kinases expressed in our heterozygous KitV558 ;T669I/+ mice, we

Stomach deduce that wild-type KIT and off-target kinase inhibition is not sufficient to affect GIST signal transduction. This apparent Δ KITV558 ;T669I-isoform dependence of phosphorylation of S6, MAPK, and STAT3 also was confirmed when Kit-mutant mice

magn. were treated with dasatinib. Treatment with dasatinib down- A* B* C* regulated phosphorylation of these components of the signal Δ transduction network in GISTs of single-mutant KitV558 /+ mice

Higher (Fig. 4De, i,andm) but not in GISTs of double-mutant Δ KitV558 ;T669I/+ mice (Fig. 4Df, j, and n). These results were confirmed by Western blotting and quantified by densitometry as the ratio of phosphorylated (pY-)S6 to total S6 protein as well as D EF pY-MAPK to total MAPK protein (Fig. 4 B and C). Importantly, Cecum the observed resistance at the histochemical and biochemical levels was mirrored in results obtained by long-term treatments: Twice daily treatment for 7 d with imatinib or dasatinib signifi- cantly reduced cell proliferation, as determined by Ki67 staining, Δ in GISTs of single-mutant KitV558 /+ mice (Fig. 4E). In contrast, G HI V558Δ;T669I/+

Colon treatment of double-mutant Kit mice with imatinib or dasatinib did not inhibit GIST proliferation, nor did it elicit a histological response (Fig. 4E and Table 1). Together, these Fig. 2. Cecal GIST and pronounced gastric and colonic ICC hyperplasia in results demonstrate that the sole addition of the gatekeeper KitV558Δ;T669I/+ mice. Cross-sections of stomach (A–C; higher magnification is mutation in an in vivo model of GIST can cause resistance, as shown in A*–C*), cecum (D–F), and colon (G–I) of 3- to 4-mo-old wild-type, postulated for patients who have imatinib-refractory GIST with

Δ Δ MEDICAL SCIENCES KitV558 /+, and KitV558 ;T669I/+ mice. Arrows indicate normal thin layer of a correlating gatekeeper mutation. myenteric ICC in wild-type samples. ICC hyperplasia in stomach and colon samples is indicated by black bars. Note the extensive hyperplasia involving Sunitinib and Sorafenib Overcome Resistance. Next, we determined V558Δ;T669I/+ Δ the circular muscle layer (dotted lines) in the stomach of Kit mice. the response of the imatinib-resistant KitV558 ;T669I/+ mice to Photographs show representative H&E staining; n ≥ 3 each. (Scale bars: second-generation TKIs, namely sunitinib and sorafenib, which 50 μminA*–C*; 100 μminA–I.) Δ hadbeenshowninvitrotoinhibitcellsexpressingtheKITV558 ;T669I mutation (17, 18). Short-term treatment with sunitinib and term treatment were evaluated. Age-matched single-mutant sorafenib reduced phosphorylation of KIT, S6, MAPK1/3, and Δ V558Δ/+ KitV558 /+ mice were treated analogously to serve as controls. STAT3 to similarly low levels in tumors of Kit and Δ V558Δ;T669I/+ To examine if the KIT kinase in KitV558 ;T669I/+ mice is sen- Kit mice as assessed by IHC (Figs. 3 C and Dk, l, q, sitive to inhibition by imatinib, tumor lysates of individual mice r, w, and x and 4Dg, h, k, l, o, and p)andWesternblotting – were subjected to Western blotting with phospho-Y719-KIT and (Figs. 3 A and B and 4 A C). After long-term treatment with Δ KIT antibodies. In tumors of imatinib-treated KitV558 /+ control sunitinib and sorafenib, KIT-mediated signal transduction and proliferation was diminished significantly in GISTs of mice, KIT phosphorylation was inhibited, as reported earlier Δ Δ V558 ;T669I/+ (Fig. 3A) (10, 11). In contrast, in tumors of KitV558 ;T669I/+ mice, Kit mice to the levels achieved by all four inhib- itors (imatinib, sunitinib, dasatinib, and sorafenib) in single- KIT phosphorylation was unchanged after treatment with ima- V558Δ/+ – – tinib (Fig. 3B). KIT inhibition could be restored by treatment mutant Kit mice (Fig. 4E and Fig. S2 Ag x and Bg x). with sunitinib, which diminished KIT phosphorylation to simi- Cell proliferation in tumor- and ICC hyperplasia-adjacent V558Δ/+ V558Δ;T669I/+ gastrointestinal epithelial cells was not impaired after long- larly low levels in both Kit and Kit mice (Fig. 3 term treatment, indicating no overt toxic side effects of these A and B). Because of the small size of neoplastic lesions in Δ TKIs in mice at the concentrations used (Fig. S3Ca–i). These KitV558 ;T669I/+ mice, the effect of drug treatment on downstream experiments demonstrated that the resistance mediated by the signaling networks was examined primarily by using an immu- V558Δ;T669I Kit mutation could be overcome by treatment with nohistochemical (IHC) approach. As is characteristic for most sunitinib and sorafenib. Δ human GIST, tumors stained positive for KIT independent To investigate whether the ICC hyperplasia in the KitV558 ;T669I/+ of the drug used or treatment duration (Figs. 3Da–f and 4Da–d – – mice recapitulated the resistance/susceptibility pattern observed and Fig. S2 Aa f and Ba f). Tumors of control (vehicle-treated) in the cecal neoplastic lesions, gastric cross-sections were ex- animals of both genotypes showed strong phosphorylation of amined after 7 d of treatment with imatinib or sunitinib. In ribosomal protein S6, MAPK, and STAT3 (Fig. 3 C and Dg, h, concordance with our results in cecal tumor lesions, the ICC Δ m, n, s, and t). After 6 h of treatment with imatinib, phosphor- hyperplasia in KitV558 ;T669I/+ mice exhibited resistance to ima- ylation of these signaling components was strongly reduced in Δ tinib and susceptibility to sunitinib inhibition of Kit signaling and tumors of KitV558 /+ animals (Fig. 3 C and Di, o, and u) but not – V558Δ;T669I ICC proliferation (Fig. S3 Cgx and D), implying that second- in tumors of animals with the Kit mutation (Fig. 3 C generation TKIs also could inhibit the early stages of imatinib- Δ and Dj, p, and v). The latter result is of particular note, because resistant GIST development in KitV558 ;T669I/+ mice. imatinib has the potential to inhibit multiple kinases (e.g., ABL and PDGFR) that might be expressed and activate these targets Hyperproliferation in Hematopoietic Cell Lineages in KitV558Δ;T669I/+ in GIST parallel to or downstream of KIT. Furthermore, in vitro Mice. In addition to the pronounced ICC hyperplasia in Δ Δ assays with the intracellular kinase domain of KIT had demon- KitV558 ;T669I/+ mice in comparison with KitV558 /+ mice, we strated that imatinib inhibits wild-type KIT as efficiently as observed hyperproliferation in other KIT-dependent lineages. Δ KITV558 (20). It is not known whether KIT heterozygosity (i.e., The mast cell hyperplasia we had observed previously in the

Bosbach et al. PNAS Early Edition | 3of8 Downloaded by guest on September 26, 2021 ACKITV558Δ/+ B KITV558Δ;T669I/+ KITV558Δ;T669I/+ V558Δ/+ V558Δ;T669I/+ ***P<0.001 Ctrl IM SU Ctrl Imatinib Sunitinib 150 500 400 pY-KIT pY-KIT 100 nuclei/field 300 + 200 50 P KIT KIT ** =0.001 100 pY-KIT / pY-KIT KIT [%] 0 0 Control Imatinib Sunitinib p-STAT3 Control Imatinib Sunitinib

D Control binitamI binitinuS V558Δ/+ V558Δ;T669I/+ V558Δ/+ V558Δ;T669I/+ V558Δ/+ V558Δ;T669I/+ abcdef KIT

ghi j kl p-S6

K q r P m n o p p-MA st uvwx p-STAT3

Δ Fig. 3. The KitV558 ;T669I double mutation confers resistance to imatinib in vivo. Sunitinib overcomes resistance. (A and B) Tumor protein lysates from in- Δ Δ dividual KitV558 /+ (A) and KitV558 ;T669I/+ (B)(Left) animals treated for 6 h with vehicle (Ctrl), imatinib (IM), or sunitinib (SU) were subjected to Western blotting Δ with phospho-Y719-KIT and KIT antibodies. Representative blots are shown. (Right) The ratio of phosphorylated to total KIT in GISTs of KitV558 ;T669I/+ mice after imatinib and sunitinib treatment was quantified by densitometry (n =3–5 animals per treatment group; error bars indicate means ± SD). (C) Quan- tification of phospho-STAT3+ nuclei per field (cf. Ds–x and corresponding explanations). n = 3 each; error bars indicate means ± SD. (D) Representative results of IHC analysis with antibodies specific for KIT (a–f), phospho-S6 ribosomal protein (g–l), phospho-MAPK (m–r), and phospho-STAT3 (s–x) on sections of GISTs from single- or double-mutant mice treated for 6 h with vehicle, imatinib, or sunitinib. Tumor sections from different treatment groups and genotypes(n ≥ 3 each) were placed next to each other on the same microscopy slide to enable cross-comparison of staining intensities. (Scale bars: 50 μm.)

Δ dorsal skin of KitV558 /+ mice (9) was exacerbated in the which becomes apparent only after 5 wk (Fig. S4Ggand h). Of Δ Δ KitV558 ;T669I/+ mice (Fig. 5C and Fig. S4Gaand b). In ap- note, the microcytic erythrocytosis of the KitV558 ;T669I/+ mice is Δ proximately half the male KitV558 ;T669I/+ mice we observed the opposite phenotype of the macrocytic associated with mast cell accumulation around Leydig cells in the interstitial Kit and Kitl loss-of-function mutations (22–24). Δ space of the testes. Interestingly, KitV558 ;T669I/+ mice ex- hibited intermittent partial alopecia of the trunk between Dependence and Pharmacological Inhibition of Erythroid Colony Growth in KitV558Δ;T669I/+ Mice. In accordance with the ele- postnatal day (P)15 and P40 (Fig. 5A), a phenotype reported Δ − − V558 ;T669I/+ for Il10 / mice and shown to be associated with increased vated hematocrit values, spleens in the Kit mice were enlarged compared with wild-type mice, and spleen cellularity of numbers of mast cells (21). Δ Δ V558 ;T669I/+ Importantly, the KitV558 ;T669I/+ mice developed a pronounced Kit mice was increased 1.5-fold without gross microcytic erythrocytosis. Manifestation of erythrocytosis first alterations in splenic architecture (Fig. 5D and Fig. S4Gcand d). was detected phenotypically by reddening of the paws, distin- Bone marrow (BM) cellularity and histology was unchanged guishable from wild-type littermates from P40 onwards (Fig. 5B). (Fig. 5D and Fig. S4Geand f). Although the erythrocytosis was reminiscent of myeloproliferative neoplasms, we noted Analysis of peripheral blood values of 8-wk-old animals revealed several marked differences: The platelet counts were not el- a substantial increase in erythrocyte counts and hematocrit val- V558Δ;T669I/+ +/+ 3 V558Δ/+ evated in Kit mice (Kit : 672 ± 248 × 10 /μLand ues in comparison with wild-type and Kit mice (Fig. S4 A V558Δ;T669I/+ 3 Kit : 681 ± 231 × 10 /μL), and there was no and B). Furthermore, the mean corpuscular volume (MCV) was myelofibrosis in either the BM or spleen, even in 1-y-old fi Δ decreased signi cantly, and the mean platelet volume was in- KITV558 ;T669I/+ mice. Kit has important functions in erythroid creased (Fig. S4 C and D). Scanning electron microscopy con- progenitors that can be revealed by the formation of in vitro burst- fi V558Δ;T669I/+ rmed that erythrocytes were smaller in diameter in Kit forming unit erythroid (BFU-E) colonies in semisolid medium. Δ mice than wild-type mice but otherwise were morphologically The number of BFU-Es obtained from the KitV558 ;T669I/+ BM normal (Fig. S4 E and F). To gain insight into the kinetics of and spleen in the presence of KitL, IL-3, and erythrocytosis and microcytosis development, we assessed he- (EPO) was higher than in wild-type mice, and a concomitant matocrit and MCV values in bleeding-naive (never before bled) increase in cfu erythroid (CFU-E) numbers also was observed in V558Δ;T669I/+ Kit mice at weeks 3, 4, 5, 6, 8, 10, 13, and 16. The the mutant BM and spleen (Table 2). KitL and EPO play a hematocrit showed a steep increase between postnatal weeks synergistic role in erythroid colony growth and survival (25). To Δ 3 and 8 (Fig. 5E). In contrast, the MCV already was reduced in test whether wild-type and KitV558 ;T669I/+ BFU-E formation 3-wk-old animals and did not change significantly over time differed in the presence of variable concentrations of , (Fig. 5F). These results indicate that the development of the BM and spleen cells were cultured in the presence of decreasing microcytosis is independent of the systemic erythrocyte overload concentrations of KitL and a fixed concentration (6 U/mL) of Δ (i.e., congestion of blood vessels in , spleen, and other organs), EPO (Fig. 6 A and B). BFU-Es from KitV558 ;T669I/+ mice

4of8 | www.pnas.org/cgi/doi/10.1073/pnas.1115240109 Bosbach et al. Downloaded by guest on September 26, 2021 PNAS PLUS A KITV558Δ;T669I/+ KITV558Δ;T669I/+ B V558Δ/+ V558Δ;T669I/+ *** P<0.001 Ctrl Dasatinib Sorafenib 150 150

100 pY-KIT 100 50 50 **P=0.004 pY-S6/S6 [%] pY-S6/S6 KIT / pY-KIT KIT [%] N/A N/A 0 0 Control Dasatinib Sorafenib Control Imatinib Sunitinib Dasatinib Sorafenib vehicledasatinibsorafenib D C **P=0.007

%] 150 [ Dasatinib Sorafenib [

V558Δ/+ V558Δ;T669I/+ V558Δ/+ V558Δ;T669I/+ 100

abcd 50

KIT 0 N/A N/A

pY-MAPK/MAPK pY-MAPK/MAPK Control Imatinib Sunitinib Dasatinib Sorafenib ef ghE After 7 days of treatment 50 p-S6 ***P<0.001 ***P<0.001

0μm field 40 i j kl 5 30

p-MAPK 20

mn o p 10 nuclei / 250μmx2 + 0 p-STAT3

Ki67 Control Imatinib Sunitinib Dasatinib Sorafenib

Δ Δ Δ Fig. 4. Resistance to dasatinib and sensitivity to sorafenib treatment in KitV558 ;T669I/+ mice. (A–C) Tumor lysates from individual KitV558 /+ and KitV558 ;T669I/+ MEDICAL SCIENCES animals treated with vehicle, imatinib, sunitinib, dasatinib, or sorafenib were subjected to Western blotting to detect the abundance of phosphorylated and total KIT, S6, and MAPK proteins. Representative blots for phospho-Y719-KIT and KIT are shown. The ratio of phosphorylated to total protein was quantified by densitometry. n =3–4 animals per treatment group; error bars indicate means ± SD. (D) Representative results after short-term treatment with dasatinib and sorafenib; IHC on GIST sections with antibodies as specified in Fig. 3D.(E) Cell proliferation in GIST of KitV558Δ;T669I/+ mice is unaffected by long-term treatment with imatinib and dasatinib. Sunitinib and sorafenib overcome resistance and attenuate cell proliferation. Ki67+ nuclei per 250 × 250 μm field; n ≥ 3 each; error bars indicate means ± SD.

showed significantly reduced dependence on KitL compared with To investigate the effect of pharmacologic intervention on the Δ Δ those from wild-type mice. The hypersensitivity of KitV558 ;T669I/+ hematopoietic phenotype in the KitV558 ;T669I/+ mice, we ana- erythroid colonies to KitL was emphasized by the observation lyzed erythroid progenitor numbers in the BM and the spleen of Δ V558Δ;T669I/+ that the BFU-Es from KitV558 ;T669I/+ BM and spleen were Kit mice after 7 d of treatment with either imatinib phenotypically larger than wild-type colonies at each concen- or sunitinib. Imatinib treatment did not change the number of Δ tration of KitL tested (Fig. 6C). Furthermore, KitV558 ;T669I/+ BFU-Es from BM and spleen compared with control vehicle- treated animals (Fig. 6 E and F); in contrast, sunitinib treatment BFU-E growth was dependent on EPO, because no colonies significantly reduced BFU-E numbers from both BM and spleen A were observed in presence of KitL alone without EPO (Fig. 6 and concomitantly reduced CFU-E numbers from both organs and B). Importantly, EPO levels in the peripheral blood were not (Fig. 6 E and F), suggesting that, in addition to the ICC hyper- fi +/+ ± V558Δ;T669I/+ signi cantly different (Kit :281.3 195 pg/mL; Kit : plasia and GIST development, the erythrocytosis phenotype of the 244.5.5 ± 146 pg/mL) (Fig. 6D), indicating that the erythrocytosis in V558Δ;T669I/+ Δ Kit mice is dependent largely on abnormal KIT the KitV558 ;T669I/+ mice was caused by increased KIT signaling in kinase activity. early erythroid progenitors. Discussion Δ By generating KITV558 ,T669I/+ mice we have produced an in V558Δ/+ Table 1. Histological response of GIST lesions in Kit and vivo model for imatinib-resistant GIST. It was unknown if the V558Δ;T669I/+ Kit mice to 7-d treatment with imatinib, dasatinib, introduction of imatinib-resistance mutations found in human sunitinib, or sorafenib GIST patients into the germ line of mice would produce viable Strain Vehicle Imatinib Dasatinib Sunitinib Sorafenib offspring or would be lethal. The heterozygous double-mutant Δ KitV558 ;T669I/+ mice we have generated not only are viable; V558Δ/+ Kit None Minimal Minimal Minimal Moderate their lifespan is extended significantly compared with that of Δ None None Minimal None Mild KitV558 /+ mice, presumably because of the reduced death rate None None Minimal Minimal Mild from intestinal obstruction by GIST. Although the gastric and Δ None colonic ICC hyperplasia is more extensive in KitV558 ;T669I/+ None V558Δ;T669I/+ mice, suggesting that the double-mutant Kit receptor is a stronger Kit None None None None Minimal oncogene, the GIST lesions in the cecum were significantly None None None None Minimal smaller, and the cecum appeared to be severely truncated. We None None None None Minimal speculate that the reduced size of the cecum may result from a Δ Minimal Minimal Minimal-mild secondary effect of ICC progenitors expressing the KITV558 ;T669I None Minimal Mild mutation on embryonic cecum development at the time of bud- None Minimal ding of the cecal pouch (26). It is of interest that the KITT670I

Bosbach et al. PNAS Early Edition | 5of8 Downloaded by guest on September 26, 2021 with previous in vitro studies (17, 18), these drugs failed to inhibit AB Δ KITV558 ;T669I/+ autophosphorylation and downstream signaling Δ in GIST lesions of the KitV558 ;T669I/+ mice. This resistance could be overcome by sunitinib and sorafenib, supporting a ra- tionale for using sunitinib as second-line therapy for imatinib- refractory GIST. Moreover, because imatinib and dasatinib V558Δ;T669I/+ +/+ V558Δ;T669I/+ inhibit wild-type KIT and other kinases in GIST, their lack of any significant impact on GIST signal transduction and pro- V558Δ;T669I/+ ** =0.003 liferation in heterozygous Kit mice underlines the C D ** =0.004 Δ dependence of tumorigenesis on the oncogenic KitV558 ;T669I 300 08 allele. Similarly, the ability of imatinib and dasatinib to inhibit <0.001 3x103×10 Δ the KIT receptor and KIT downstream signaling in KitV558 /+ V558Δ;T669I/+ 200 2x102×1008 but not in Kit mice confirms that the response to Δ imatinib and dasatinib treatment in KitV558 /+ mice is mostly 08

100 Nucleated cells 1x101×10 the result of direct inhibition of oncogenic KIT signaling rather Mast cells / cells / cm Mast than off-target effects. These observations underline the useful- Δ Δ 0 0 ness of the KitV558 /+ and KitV558 ;T669I/+ mice in investigating +/+ V558Δ/+ GTK +/+ GTK +/+ GTK oncogenic KIT signaling in GIST and their utility for preclinical BM Spleen E F studies of drug candidates. 70 V558Δ; T669I/+ 70 The Kit receptor has a critical role in erythropoiesis and mast 60 60 cells, and Kit loss-of-function mutations result in macrocytic anemia and mast cell deficiency (22, 28). Mice with the onco- Δ 50 50 +/+ genic KITV558 mutation, in addition to GIST, develop mast cell 40 +/+ 40 hyperplasia, but peripheral blood values are normal (9). In the V558Δ;T669I/+ Hematocrit [%] Kit mice the mast cell hyperplasia is increased, but, 30 30 V558Δ;T669I/+ Mean Cell Volume [fL] fascinatingly, these mice develop a highly penetrant eryth- 20 20 rocytosis similar to that seen in myeloproliferative neoplasms 0 5 10 15 0 5 10 15 and polycythemia vera (PV) in addition to GIST. KIT gain-of- ]skeew[egA ]skeew[egA function mutations have been identified in mastocytosis and in Fig. 5. Increased mast cell and numbers in KitV558Δ;T669I/+ acute myelogenous leukemia but not in PV, although KIT mice. (A) Representative photograph of alopecia of truncal hair in 4-wk-old mutations of unclear significance have been reported in PV KitV558Δ;T669I/+ mice. (B) Representative photographs of wild-type hind-paw previously (29). Therefore it was somewhat surprising that Δ Δ in 3-mo-old wild-type mice and “red paw” phenotype in 3-mo-old KitV558 ;T669I/+ KitV558 ;T669I/+ mice developed erythrocytosis. Because KIT is Δ mice. (C) Increased mast cell numbers in dorsal skin of KitV558 ;T669I/+ (GTK) Δ known to have a role in early erythroid progenitors, the hyper- mice in comparison with wild-type and KitV558 /+ mice (n =3).(D)BMcellularity V558Δ;T669I/+ sensitivity observed in BFU-E assays and the lack of an effect on of Kit mice measured as total nucleated cells per bone is similar to later progenitors and on serum erythropoietin levels confirms the that in wild-type mice. Spleen cellularity is significantly increased in the mutants compared with wild-type mice. n =6.(E) Time course of hematocrit role of KIT in early erythropoiesis (25). These observations also and (F) MCV showing development of erythrocytosis and microcytosis phe- serve to distinguish PV (the molecular basis of which most often is Δ notype in KitV558 ;T669I/+ mice (red curves) in and wild-type mice (black curves). an activating mutation in JAK2) from the microcytic erythrocytosis V558Δ;T669I/+ n =5–17. Error bars indicate means ± SD. in the Kit mice (30). The erythroid phenotype in the Δ KitV558 ;T669I/+ mice becomes evident between 3 and 5 wk of age. Interestingly, the earliest manifestation of erythrocytosis in the Δ mutation is found only in imatinib-resistant GIST patients in KitV558 ;T669I/+ mice coincides with the switch from fetal to adult combination with a primary KIT mutation. There are no reports hematopoiesis at 3 wk of age, and this timing may reflect the to indicate that the KITT670I mutation may be found alone as a distinct cellular contexts in fetal and adult hematopoiesis (31). Δ Δ single-site mutation and on its own may have a role in GIST In comparison with KitV558 ,theKitV558 ;T669I mutation in vivo tumorigenesis, although in vitro experiments in COS cells with produces increased ICC hyperplasia and more pronounced, as T670I well as distinct, hematopoietic phenotypes. Although in vitro the single-mutant KIT in the absence of KitL indicated KIT Δ autoactivation of the KITV558 kinase is unchanged compared autophosphorylation (27). V558Δ;T669I V558Δ/+ V558Δ;T669I/+ with the KIT kinase, the kinase activity of the Although GISTs of Kit and Kit mice were Δ Δ V558 ;T669I similar in histology and oncogenic signaling, the KitV558 ;T669I/+ KIT kinase is nearly doubled and could explain the KIT hyperactivity observed in the ICC, mast cell, and erythro- mice were resistant to imatinib and dasatinib therapy. In agreement Δ cyte lineages in KitV558 ;T669I/+ mice (32). However, at this time we cannot rule out the possibility that the T669I second-site Δ mutation alters oncogenic Kit signaling in a qualitative fashion. Table 2. Increased erythroid progenitors in spleen of KitV558 ;T669I/+ Our results highlight the importance of a combination of factors, mice including the type of activation mutation and the cellular con- BM Spleen text, in determining mutant/oncogenic phenotypes in vivo. Colony +/+ V558Δ;T669I/+ +/+ V558Δ;T669I/+ type KIT Kit KIT Kit Materials and Methods Δ BFU-E 3,938 ± 1,042 6,386 ± 3,193 3,317 ± 1,436 16,198 ± 8,439 Generation of Mouse Strains. The V558 mutation was introduced by CFU-E 1,707 ± 910 4,562 ± 1,645 3,731 ± 1,759 25,575 ± 9,645* site-directed mutagenesis into a 2.1-kb EcoRI/MluI fragment across Kit exons 9–11 from a 129/SvJ mouse library (19) serving as the 5′ homology arm Δ BM and spleen cells from wild-type and KitV558 ;T669I/+ mice were cultured of the targeting vector (Fig.1A). It was linked 3′ to a floxed neomycin- in the presence of KitL (100 ng/mL), EPO (6 U/mL), and IL-3 (100 ng/mL). CFU-E resistance (NEO) gene-expression cassette. The 3′ homology arm for the Δ and BFU-E colonies were scored at the end of 2 and 10 d, respectively. n =3 KitV558 allele was a 1.3-kb MluI/BsrGI-NcoI fragment across exons 12–13 Δ mice per group. (Fig. S1A). For the KitV558 ;T669I allele this arm was elongated by a 2.1-kb BsrGI- *P < 0.05 in comparison with wild type. NcoI/BamHI fragment including the T669I mutation in exon 14. To improve

6of8 | www.pnas.org/cgi/doi/10.1073/pnas.1115240109 Bosbach et al. Downloaded by guest on September 26, 2021 the C57BL/6J background. A PCR strategy bracketing the 134-bp loxP scar was PNAS PLUS A 60 * =0.04 +/+ B 50 *** 0.001 V558Δ;T669I/+ used for routine genotyping thereafter (wild-type allele, 291 bp; targeted al- * =0.04 +/+ lele, 425 bp) with the primers mKitEx11F, 5′-CATAGACCCGACGCAACTTC-3′; 40 V558Δ;T669I/+ mKitIn11R, 5′-GGTTCCCAAATCAACAAGGC-3′. Initial experiments were done 40 ** =0.009 30 * =0.04 *** 0.001 with backcross generation N4, and repetitions were done with N6–N13 ani- mals. No change in phenotype was observed in different backcross gen- 20 erations. All animal procedures were approved by the Institutional Animal 20 Care and Use Committee of Memorial Sloan Kettering Cancer Center. 10 BFU-E / 50,000 BM cells BFU-E / 100,000 spleen cells / BFU-E Drug Treatments. Cohorts of 3- to 4-mo-old animals were injected i.p. twice 0 0 daily with imatinib mesylate (45 mg/kg body mass) (Novartis), dasatinib salt EPO + + + + - EPO + + + + - (25 mg/kg; synthesized in house), or vehicle control (water). Sunitinib ma- KitL [ng] 20 5 1 - 20 KitL [ng] 20 E 5E 1 - 20 + + late (40 mg/kg) or sorafenib tosylate (60 mg/kg) (both from LC Laboratories) C D were solubilized freshly in 30% (vol/vol) Cremophor EL (Sigma Aldrich), 30% +/+ V558Δ;T669I/+ (vol/vol) polyethylene glycol 400, 10% (vol/vol) ethanol, and 10% (wt/vol) 600 glucose every 4 d and were administered once daily by gavage.

Histologic and IHC Analyses. Microscopic and IHC analyses were performed as 400 =n.s. described previously (10). Animals were dissected 6 h after injection [sor-

20 ng KitL+EPO afenib-treated animals were dissected 3 h after injection (34)], and tumors were frozen immediately in liquid nitrogen and/or were fixed in fresh 4%

pg/mL 200 a EPO conc. [pg/ml] PO paraformaldehyde at 4 °C overnight. After paraffin embedding, 5-μmsec- E m tions of the tumors of at least three different animals per treatment group

Plas and of at least three different treatments were placed on microscopic slides 0 next to one another to enable cross-comparison within a slide after IHC 1 ng KitL+ +/+ V558Δ;T669I/+ staining with the antibodies indicated. Antibodies used were Phospho E F (P)-ribosomal protein S6 (S235/236) (D57.2.2E), P-MAPK (ERK-1/2) (T202/ 15000 100000 Δ; Control =0.016 8 Control 204) (20G11), and P-STAT3 (Y705) (D3A7) (Cell Signaling); Ki67 (Vector 5 =n.s. 5 Imatinib V Imatinib Laboratories); and KIT (Oncogene Science). Stained slides were scanned 80000 it Sunitinib K Sunitinib with MIRAX Scan (Carl Zeiss), and Ki67 and STAT3 phosphorylation were 10000 =0.054 quantified by counting stained nuclei in nine 250 × 250 μm fields of at least 60000 MEDICAL SCIENCES =0.043 three different tumors per genotype and treatment. en or femur en or en or femur en or e e =n.s. 40000 =n.s. Tissues for H&E staining were fixed in 10% neutral buffered formalin, and 5000 * =0.025 =0.064 bones were decalcified in 0.5 M EDTA. Tissue samples were subjected to 20000 routine histological procedures and embedded in paraffin. Five-micrometer BFU-E / spl CFU-E / spl sections were used for histological staining with H&E. 0 0 BFU-E/spleen BFU-E/femur CFU-E/spleen CFU-E/femur Western Blotting. Tumor lysates and Western blotting with KIT (D13A2), P-KIT Δ Fig. 6. KitV558 ;T669I/+ erythroid progenitor growth is hypersensitive to KitL (Y719), S6 (5G10), P-S6 (S235/236) (D57.2.2E), MAPK (9102), and P-MAPK (ERK-1/2) and is susceptible to sunitinib inhibition. (A and B)BM(A) and spleen (B) cells (T202/204) (20G11) antibodies (all from Cell Signaling) were done as described fi were cultured in colony assays in the presence of EPO and decreasing levels (9) with the following modi cations. Liquid nitrogen-frozen samples were Δ μ of KitL. KitV558 ;T669I/+ BM gives rise to a significantly higher number of minced in 50 L lysis buffer in Petri dishes on solid CO2 with scalpels, sonicated Δ fi BFU-Es for each KitL+EPO combination. KitV558 ;T669I/+ spleen also gives rise twice for 10 s with a Soni er S-250A with double-stepped microtip (Branson to significantly greater number of BFU-Es at all the KitL+EPO combinations Ultrasonics), and cleared by centrifugation. Densitometry of Western blot- fi tested compared with wild-type spleen. Error bars indicate means ± SD; ting bands was done with Multi Gauge V3.1 (Fuji lm) by setting the ratio of n =3.(C) Representative images depict size of BFU-E colonies obtained from the intensities of control-treated phospho-KIT bands and their correspond- wild-type and KitV558Δ;T669I/+ BM. (Magnification: 10 ×.) (D) EPO levels from ing total KIT bands to 100% and normalizing the ratio of the treated sera of wild-type and KitV558Δ;T669I/+ mice did not differ statistically. Error bars phospho-KIT/total KIT band intensities to that standard. Δ indicate means ± SD. (E and F) KitV558 ;T669I/+ mice were treated with vehicle, imatinib, or sunitinib for 7 d. Femoral BM and spleen cells were cultured in Colony-Forming Assay. Unfractionated BM cells (50,000 cells per plate) or the presence of KitL (100 ng/mL), EPO (6 U/mL), and IL3 (100 ng/mL). BFU-E spleen cells (100,000 cells per plate) from at least three mice were plated in (E) and CFU-E (F) colonies were scored after 10 d and 3 d, respectively. Error 1 mL Iscove’s modified Dulbecco’s medium containing 1.2% (wt/vol) meth- bars indicate means ± SD; n =3. ylcellulose, 30% FCS, 2 mmol/L glutamine, 0.1 mmol/L 2-mercaptoethanol, and 4 mmol/L hemin with cytokines (EPO 6 U/mL; IL-3 100 ng/mL, and KitL 100 ng/mL). Cultures were maintained in triplicate at 37 °C in humidified 5%

homologous recombination, a 3.8-kb BamHI fragment across exons 15–17 CO2. CFU-Es were scored after 2 d, and BFU-E colonies were scored after was added 3′ of T669I. The final targeting vector was sequenced completely 7–10 d of culture. BFU-E growth also was assessed in the presence of 6 U/mL before linearization and electroporation into CJ7 ES cells (33). Screening of EPO and 1, 5, or 20 ng/mL KitL. Colonies were scored using Nikon Eclipse TE 192 clones by Southern blot with a 5′ external probe across Kit exon 8 200, and images were taken Nikon Eclipse Ti-S. yielded four positive clones (Fig. 1A and Fig. S1B). One carried both the V558Δ;T669I/+ V558Δ mutation and the T669I mutation as assessed by sequencing and Erythropoietin Measurement. Sera from wild-type or Kit mice showed a normal karyotype. In C57BL/6J blastocyst injections this clone gave obtained by retroorbital bleeding were used to measure EPO levels using ’ rise to 11 high-grade chimeras (more than 90% agouti coat color). After the Milliplex assay system (Millipore) according to the manufacturer s crossing to C57BL/6J mice, in all agouti F1 animals heterozygous for the NEO instructions. allele, the presence of the V558Δ and T669I mutations and the integrity of both loxP sites were confirmed by sequencing. To remove the floxed Statistical Analysis. Comparison between two groups was done by unpaired t Δ fi NEO cassette, F1 KitV558 ;T669I-NEO/+ males were bred to B6.FVB-Tg(EIIa-cre) test analysis using GraphPad Prism (version 5.0). Statistical signi cance was < C5379Lmgd/J females (Jackson Laboratory). Genotyping PCR was performed achieved when P 0.05. across the original intron 11 MluI site, which was replaced by a 134-bp loxP scar in the case of the targeted alleles. Of the two resulting fragments [wild-type ACKNOWLEDGMENTS. We thank Maureen Sullivan, LingBo Shen, Yasemin Yozgat, and Fabienne Brenet for expert assistance; Aleksander M. Baldys, allele (643 bp) and targeted allele (777 bp)], the longer was isolated by gel Afsar Barlas, Ning Fan, Volodia Gueorguiev, and Mesruh Turkekul from the fi Δ puri cation, and the integrity of the V558 mutation and of the remaining Molecular Cytology, Huiyong Zhao from the Antitumor Assessment, and V558Δ;T669I loxP site was confirmed by sequencing (Fig. S1C). Only Kit animals that Willie H. Mark and Antoinette Rookard from the Mouse Genetics Core genotyped negative for the Cre allele were used for subsequent backcrosses to Facilities at Memorial Sloan-Kettering Cancer Center, and Chingwen Yang

Bosbach et al. PNAS Early Edition | 7of8 Downloaded by guest on September 26, 2021 from the Rockefeller University Gene Targeting Resource Center for expert CA102774, RO1-HL55748, and P50-CA140146 (to P.B.) and by grants from help; and John Burrowes and Zachary Oberzan for secretarial assistance. the Starr Cancer Consortium (to P.B., C.R.A., and J.M.S.) and the LifeRaft This work was supported by National Institutes of Health Grants R01- Group (to P.B. and C.R.A.).

1. Kwon JG, et al. (2009) Changes in the structure and function of ICC networks in 18. Guo T, et al. (2007) Sorafenib inhibits the imatinib-resistant KITT670I gatekeeper ICC hyperplasia and gastrointestinal stromal tumors. Gastroenterology 136: mutation in gastrointestinal stromal tumor. Clin Cancer Res 13:4874–4881. 630–639. 19. Kissel H, et al. (2000) Point mutation in reveals essential 2. Heinrich MC, et al. (2003) PDGFRA activating mutations in gastrointestinal stromal roles for kit signaling in spermatogenesis and oogenesis without affecting other kit tumors. Science 299:708–710. responses. EMBO J 19:1312–1326. 3. Hirota S, et al. (1998) Gain-of-function mutations of c-kit in human gastrointestinal 20. Karaman MW, et al. (2008) A quantitative analysis of kinase inhibitor selectivity. Nat stromal tumors. Science 279:577–580. Biotechnol 26:127–132. 4. Antonescu CR, et al. (2003) Association of KIT exon 9 mutations with nongastric 21. Vanderford DA, et al. (2010) Alopecia in IL-10-deficient mouse pups is c-kit-dependent primary site and aggressive behavior: KIT mutation analysis and clinical correlates of and can be triggered by iron deficiency. Exp Dermatol 19:518–526. 120 gastrointestinal stromal tumors. Clin Cancer Res 9:3329–3337. 22. Russell ES (1979) Hereditary of the mouse: A review for geneticists. Adv 5. Lasota J, et al. (2000) Mutations in exons 9 and 13 of KIT gene are rare events in Genet 20:357–459. – gastrointestinal stromal tumors. A study of 200 cases. Am J Pathol 157:1091 1095. 23. Bernstein A, et al. (1990) The mouse W/c-kit locus. Ciba Found Symp 148:158–166, 6. Rubin BP, et al. (2001) KIT activation is a ubiquitous feature of gastrointestinal discussion 166–172. – stromal tumors. Cancer Res 61:8118 8121. 24. Besmer P (1997) Kit-Ligand- (Marcel Dekker, New York, NY), pp 7. Mol CD, et al. (2004) Structural basis for the autoinhibition and STI-571 inhibition of 369–403. – c-Kit tyrosine kinase. J Biol Chem 279:31655 31663. 25. Nocka K, Buck J, Levi E, Besmer P (1990) Candidate ligand for the c-kit transmembrane 8. Nishida T, et al. (1998) Familial gastrointestinal stromal tumours with germline kinase receptor: KL, a fibroblast derived stimulates mast cells and mutation of the KIT gene. Nat Genet 19:323–324. erythroid progenitors. EMBO J 9:3287–3294. 9. Sommer G, et al. (2003) Gastrointestinal stromal tumors in a mouse model by 26. Zhang X, et al. (2006) Reciprocal epithelial-mesenchymal FGF signaling is required for targeted mutation of the Kit receptor tyrosine kinase. Proc Natl Acad Sci USA 100: cecal development. Development 133:173–180. 6706–6711. 27. Tamborini E, et al. (2006) Functional analyses and molecular modeling of two c-Kit 10. Rossi F, et al. (2006) Oncogenic Kit signaling and therapeutic intervention in a mutations responsible for imatinib secondary resistance in GIST patients. Oncogene mouse model of gastrointestinal stromal tumor. Proc Natl Acad Sci USA 103: 25:6140–6146. 12843–12848. 28. Kitamura Y, Go S, Hatanaka K (1978) Decrease of mast cells in W/Wv mice and their 11. Rossi F, et al. (2010) Imatinib upregulates compensatory signaling in a mouse increase by bone marrow transplantation. Blood 52:447–452. model of gastrointestinal stromal tumor and is more effective when combined with 29. Fontalba A, et al. (2006) Identification of c-Kit gene mutations in patients with dasatinib. Mol Cancer Res 8:1271–1283. polycythemia vera. Leuk Res 30:1325–1326. 12. Antonescu CR, et al. (2005) Acquired resistance to imatinib in gastrointestinal stromal tumor occurs through secondary gene mutation. Clin Cancer Res 11:4182–4190. 30. Levine RL, Pardanani A, Tefferi A, Gilliland DG (2007) Role of JAK2 in the – 13. Debiec-Rychter M, et al. (2005) Mechanisms of resistance to imatinib mesylate in pathogenesis and therapy of myeloproliferative disorders. Nat Rev Cancer 7:673 683. fi gastrointestinal stromal tumors and activity of the PKC412 inhibitor against imatinib- 31. Bowie MB, et al. (2007) Identi cation of a new intrinsically timed developmental resistant mutants. Gastroenterology 128:270–279. checkpoint that reprograms key hematopoietic stem cell properties. Proc Natl Acad – 14. Gorre ME, et al. (2001) Clinical resistance to STI-571 cancer therapy caused by BCR-ABL Sci USA 104:5878 5882. gene mutation or amplification. Science 293:876–880. 32. Gajiwala KS, et al. (2009) KIT kinase mutants show unique mechanisms of drug 15. Shah NP, et al. (2002) Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to imatinib and sunitinib in gastrointestinal stromal tumor patients. Proc resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast Natl Acad Sci USA 106:1542–1547. crisis chronic myeloid leukemia. Cancer Cell 2:117–125. 33. Swiatek PJ, Gridley T (1993) Perinatal lethality and defects in hindbrain development 16. Tamborini E, et al. (2004) A new mutation in the KIT ATP pocket causes acquired in mice homozygous for a targeted mutation of the zinc finger gene Krox20. Genes resistance to imatinib in a gastrointestinal stromal tumor patient. Gastroenterology Dev 7:2071–2084. 127:294–299. 34. Liu L, et al. (2006) Sorafenib blocks the RAF/MEK/ERK pathway, inhibits tumor 17. Carter TA, et al. (2005) Inhibition of drug-resistant mutants of ABL, KIT, and EGF angiogenesis, and induces tumor cell apoptosis in model receptor kinases. Proc Natl Acad Sci USA 102:11011–11016. PLC/PRF/5. Cancer Res 66:11851–11858.

8of8 | www.pnas.org/cgi/doi/10.1073/pnas.1115240109 Bosbach et al. Downloaded by guest on September 26, 2021