Compressed phenotypic screens for complex multicellular models and high-content assays.

High-throughput phenotypic screens leveraging biochemical perturbations, high-content readouts, and complex multicellular models could advance therapeutic discovery yet remain constrained by limitations of scale. To address this, we establish a method for compressing screens by pooling perturbations followed by computational deconvolution. Conducting controlled benchmarks with a highly bioactive small molecule library and a high-content imaging readout, we demonstrate increased efficiency for compressed experimental designs compared to conventional approaches. To prove generalizability, we apply compressed screening to examine transcriptional responses of patient-derived pancreatic cancer organoids to a library of tumor-microenvironment (TME)-nominated recombinant protein ligands. Using single-cell RNA-seq as a readout, we uncover reproducible phenotypic shifts induced by ligands that correlate with clinical features in larger datasets and are distinct from reference signatures available in public databases. In sum, our approach enables phenotypic screens that interrogate complex multicellular models with rich phenotypic readouts to advance translatable drug discovery as well as basic biology.

Glycolysis regulates KRAS plasma membrane localization and function through defined glycosphingolipids.

Oncogenic KRAS expression generates a metabolic dependency on aerobic glycolysis, known as the Warburg effect. We report an effect of increased glycolytic flux that feeds into glycosphingolipid biosynthesis and is directly linked to KRAS oncogenic function. High resolution imaging and genetic approaches show that a defined subset of outer leaflet glycosphingolipids, including GM3 and SM4, is required to maintain KRAS plasma membrane localization, with GM3 engaging in cross-bilayer coupling to maintain inner leaflet phosphatidylserine content. Thus, glycolysis is critical for KRAS plasma membrane localization and nanoscale spatial organization. Reciprocally oncogenic KRAS selectively upregulates cellular content of these same glycosphingolipids, whose depletion in turn abrogates KRAS oncogenesis in pancreatic cancer models. Our findings expand the role of the Warburg effect beyond ATP generation and biomass building to high-level regulation of KRAS function. The positive feedforward loop between oncogenic KRAS signaling and glycosphingolipid synthesis represents a vulnerability with therapeutic potential.

Components of the phosphatidylserine endoplasmic reticulum to plasma membrane transport mechanism as targets for KRAS inhibition in pancreatic cancer.

KRAS is mutated in 90% of human pancreatic ductal adenocarcinomas (PDACs). To function, KRAS must localize to the plasma membrane (PM) via a C-terminal membrane anchor that specifically engages phosphatidylserine (PtdSer). This anchor-binding specificity renders KRAS-PM localization and signaling capacity critically dependent on PM PtdSer content. We now show that the PtdSer lipid transport proteins, ORP5 and ORP8, which are essential for maintaining PM PtdSer levels and hence KRAS PM localization, are required for KRAS oncogenesis. Knockdown of either protein, separately or simultaneously, abrogated growth of KRAS-mutant but not KRAS-wild-type pancreatic cancer cell xenografts. ORP5 or ORP8 knockout also abrogated tumor growth in an immune-competent orthotopic pancreatic cancer mouse model. Analysis of human datasets revealed that all components of this PtdSer transport mechanism, including the PM-localized EFR3A-PI4KIIIα complex that generates phosphatidylinositol-4-phosphate (PI4P), and endoplasmic reticulum (ER)-localized SAC1 phosphatase that hydrolyzes counter transported PI4P, are significantly up-regulated in pancreatic tumors compared to normal tissue. Taken together, these results support targeting PI4KIIIα in KRAS-mutant cancers to deplete the PM-to-ER PI4P gradient, reducing PM PtdSer content. We therefore repurposed the US Food and Drug Administration-approved hepatitis C antiviral agent, simeprevir, as a PI4KIIIα inhibitor In a PDAC setting. Simeprevir potently mislocalized KRAS from the PM, reduced the clonogenic potential of pancreatic cancer cell lines in vitro, and abrogated the growth of KRAS-dependent tumors in vivo with enhanced efficacy when combined with MAPK and PI3K inhibitors. We conclude that the cellular ER-to-PM PtdSer transport mechanism is essential for KRAS PM localization and oncogenesis and is accessible to therapeutic intervention.

Oncogenic KRAS is dependent upon an EFR3A-PI4KA signaling axis for potent tumorigenic activity.

The HRAS, NRAS, and KRAS genes are collectively mutated in a fifth of all human cancers. These mutations render RAS GTP-bound and active, constitutively binding effector proteins to promote signaling conducive to tumorigenic growth. To further elucidate how RAS oncoproteins signal, we mined RAS interactomes for potential vulnerabilities. Here we identify EFR3A, an adapter protein for the phosphatidylinositol kinase PI4KA, to preferentially bind oncogenic KRAS. Disrupting EFR3A or PI4KA reduces phosphatidylinositol-4-phosphate, phosphatidylserine, and KRAS levels at the plasma membrane, as well as oncogenic signaling and tumorigenesis, phenotypes rescued by tethering PI4KA to the plasma membrane. Finally, we show that a selective PI4KA inhibitor augments the antineoplastic activity of the KRASG12C inhibitor sotorasib, suggesting a clinical path to exploit this pathway. In sum, we have discovered a distinct KRAS signaling axis with actionable therapeutic potential for the treatment of KRAS-mutant cancers.

RAS Function in cancer cells: translating membrane biology and biochemistry into new therapeutics.

The three human RAS proteins are mutated and constitutively activated in ∼20% of cancers leading to cell growth and proliferation. For the past three decades, many attempts have been made to inhibit these proteins with little success. Recently; however, multiple methods have emerged to inhibit KRAS, the most prevalently mutated isoform. These methods and the underlying biology will be discussed in this review with a special focus on KRAS-plasma membrane interactions.

Targeting plasma membrane phosphatidylserine content to inhibit oncogenic KRAS function.

The small GTPase KRAS, which is frequently mutated in human cancers, must be localized to the plasma membrane (PM) for biological activity. We recently showed that the KRAS C-terminal membrane anchor exhibits exquisite lipid-binding specificity for select species of phosphatidylserine (PtdSer). We, therefore, investigated whether reducing PM PtdSer content is sufficient to abrogate KRAS oncogenesis. Oxysterol-related binding proteins ORP5 and ORP8 exchange PtdSer synthesized in the ER for phosphatidyl-4-phosphate synthesized in the PM. We show that depletion of ORP5 or ORP8 reduced PM PtdSer levels, resulting in extensive mislocalization of KRAS from the PM. Concordantly, ORP5 or ORP8 depletion significantly reduced proliferation and anchorage-independent growth of multiple KRAS-dependent cancer cell lines, and attenuated KRAS signaling in vivo. Similarly, functionally inhibiting ORP5 and ORP8 by inhibiting PI4KIIIα-mediated synthesis of phosphatidyl-4-phosphate at the PM selectively inhibited the growth of KRAS-dependent cancer cell lines over normal cells. Inhibiting KRAS function through regulating PM lipid PtdSer content may represent a viable strategy for KRAS-driven cancers.

Acylpeptide hydrolase is a novel regulator of KRAS plasma membrane localization and function.

The primary site for KRAS signaling is the inner leaflet of the plasma membrane (PM). We previously reported that oxanthroquinone G01 (G01) inhibited KRAS PM localization and blocked KRAS signaling. In this study, we identified acylpeptide hydrolase (APEH) as a molecular target of G01. APEH formed a stable complex with biotinylated G01, and the enzymatic activity of APEH was inhibited by G01. APEH knockdown caused profound mislocalization of KRAS and reduced clustering of KRAS that remained PM localized. APEH knockdown also disrupted the PM localization of phosphatidylserine (PtdSer), a lipid critical for KRAS PM binding and clustering. The mislocalization of KRAS was fully rescued by ectopic expression of APEH in knockdown cells. APEH knockdown disrupted the endocytic recycling of epidermal growth factor receptor and transferrin receptor, suggesting that abrogation of recycling endosome function was mechanistically linked to the loss of KRAS and PtdSer from the PM. APEH knockdown abrogated RAS-RAF-MAPK signaling in cells expressing the constitutively active (oncogenic) mutant of KRAS (KRASG12V), and selectively inhibited the proliferation of KRAS-transformed pancreatic cancer cells. Taken together, these results identify APEH as a novel drug target for a potential anti-KRAS therapeutic.

Molecular Analysis of Congenital Hypothyroidism in Saudi Arabia: SLC26A7 Mutation Is a Novel Defect in Thyroid Dyshormonogenesis.

Context: Congenital hypothyroidism (CH) is the most common neonatal endocrine disorder, affecting one in 3000 to 4000 newborns. Since the introduction of a newborn screening program in 1988, more than 300 cases have been identified. The underlying genetic defects have not been systematically studied. Objective: To identify the mutation spectrum of CH-causing genes. Methods: Fifty-five patients from 47 families were studied by next-generation exome sequencing. Results: Mutations were identified in 52.7% of patients (29 of 55) in the following 11 genes: TG, TPO, DUOX2, SLC26A4, SLC26A7, TSHB, TSHR, NKX2-1, PAX8, CDCA8, and HOXB3. Among 30 patients with thyroid dyshormonogenesis, biallelic TG mutations were found in 12 patients (40%), followed by biallelic mutations in TPO (6.7%), SLC26A7 (6.7%), and DUOX2 (3.3%). Monoallelic SLC26A4 mutations were found in two patients, one of them coexisting with two tandem biallelic deletions in SLC26A7. In 25 patients with thyroid dysgenesis, biallelic mutations in TSHR were found in six patients (24%). Biallelic mutations in TSHB, PAX 8, NKX2-1, or HOXB3 were found once in four different patients. A monoallelic CDCA8 mutation was found in one patient. Most mutations were novel, including three TG, two TSHR, and one each in DUOX2, TPO, SLC26A7, TSHB, NKX2-1, PAX8, CDCA8, and HOXB3. SLC26A7 and HOXB3 were novel genes associated with thyroid dyshormonogenesis and dysgenesis, respectively. Conclusions: TG and TSHR mutations are the most common genetic defects in Saudi patients with CH. The prevalence of other disease-causing mutations is low, reflecting the consanguineous nature of the population. SLC26A7 mutations appear to be associated with thyroid dyshormonogenesis.

Clinical and genetic characteristics of 15 families with hereditary hypophosphatemia: Novel Mutations in PHEX and SLC34A3.

BACKGROUND: Hereditary hypophosphatemia is a group of rare renal phosphate wasting disorders. The diagnosis is based on clinical, radiological, and biochemical features, and may require genetic testing to be confirmed. METHODOLOGY: Clinical features and mutation spectrum were investigated in patients with hereditary hypophosphatemia. Genomic DNA of 23 patients from 15 unrelated families were screened sequentially by PCR-sequencing analysis for mutations in the following genes: PHEX, FGF23, DMP1, ENPP1, CLCN5, SLC34A3 and SLC34A1. CytoScan HD Array was used to identify large deletions. RESULTS: Genetic evaluation resulted in the identification of an additional asymptomatic but intermittent hypophosphatemic subject. Mutations were detected in 21 patients and an asymptomatic sibling from 13 families (86.6%, 13/15). PHEX mutations were identified in 20 patients from 12 families. Six of them were novel mutations present in 9 patients: c.983_987dupCTACC, c.1586+2T>G, c.1206delA, c.436+1G>T, c.1217G>T, and g.22,215,887-22,395,767del (179880 bp deletion including exon 16-22 and ZNF645). Six previously reported mutations were found in 11 patients. Among 12 different PHEX mutations, 6 were de novo mutations. Patients with de novo PHEX mutations often had delayed diagnosis and significantly shorter in height than those who had inherited PHEX mutations. Novel compound heterozygous mutations in SLC34A3 were found in one patient and his asymptomatic sister: c.1335+2T>A and c.1639_1652del14. No mutation was detected in two families. CONCLUSIONS: This is the largest familial study on Turkish patients with hereditary hypophosphatemia. PHEX mutations, including various novel and de novo variants, are the most common genetic defect. More attention should be paid to hypophosphatemia by clinicians since some cases remain undiagnosed both during childhood and adulthood.

Mutational analysis of PHEX, FGF23 and CLCN5 in patients with hypophosphataemic rickets.

CONTEXT: Hypophosphataemic rickets (HR) is a group of rare hereditary renal phosphate wasting disorders caused by mutations in PHEX, FGF23, DMP1, ENPP1, CLCN5 or SLC34A3. OBJECTIVE: To investigate underlying genetic defects in patients with hypophosphataemic rickets. METHODS: We analysed genomic DNA from nine unrelated families for mutations in the entire coding region of PHEX, FGF23, DMP1, ENPP1, CLCN5 or SLC34A3 by PCR sequencing and copy number analysis. RESULTS: A total of 14 patients were studied. PHEX mutations were identified in 12 patients from seven families. Five of them were novel mutations present in eight patients: c.154G>T (p.E52*), c.401_402insGCCAAA (p.Q134_K135insPK), c.1600C>T (p.P534S), g.22016715_22056805del (40-kb deletion including promoter and exons 1-3) and c.2242_2243delCT (p.L748 fs*48). Four patients had previously reported mutations: c.1768+1G>A and c.1807G>A (p.W602*). Novel CLCN5 (c.1205G>A, p.W402*) and FGF23 (c.526C>G, p.R176G) mutations were found in two patients from the remaining two families. Many of the mutations were de novo: c.154G>T and c.2242_2243delCT in PHEX and c.526C>G in FGF23. Furthermore, we characterized the breakpoint of the novel PHEX g.22016715_22056805del and the c.2242_2243delCT, which is 6 bp from the stop codon, resulting in a frameshift and extension of the reading frame by 42 amino acids. CONCLUSIONS: Novel and de novo mutations are frequent and PHEX mutations are still the most common genetic defects in the Turkish population. Gene copy number analysis should be considered in patients with negative results by conventional PCR-based sequencing analysis. The current study further expands the mutation spectrum underlying HR.

Novel CYP27B1 Gene Mutations in Patients with Vitamin D-Dependent Rickets Type 1A.

The CYP27B1 gene encodes 25-hydroxyvitamin D-1α-hydroxylase. Mutations of this gene cause vitamin D-dependent rickets type 1A (VDDR-IA, OMIM 264700), which is a rare autosomal recessive disorder. To investigate CYP27B1 mutations, we studied 8 patients from 7 unrelated families. All coding exons and intron-exon boundaries of CYP27B1 gene were amplified by PCR from peripheral leukocyte DNA and subsequently sequenced. Homozygous mutations in the CYP27B1 gene were found in all the patients and heterozygous mutations were present in their normal parents. One novel single nucleotide variation (SNV, c.1215 T>C, p.R379R in the last nucleotide of exon 7) and three novel mutations were identified:, a splice donor site mutation (c.1215+2T>A) in intron 7, a 16-bp deletion in exon 6 (c.1022-1037del16), and a 2-bp deletion in exon 5 (c.934_935delAC). Both c.1215 T>C and c.1215+2T>A were present together in homozygous form in two unrelated patients, and caused exon 7 skipping. However, c.1215 T>C alone has no effect on pre-mRNA splicing. The skipping of exon 7 resulted in a shift of downstream reading frame and a premature stop codon 57 amino acids from L380 (p.L380Afs*57). The intra-exon deletions of c.1022-1037del16 and c.934_935delAC also resulted in a frameshift and the creation of premature stop codons at p.T341Rfs*5, and p.T312Rfs*19, respectively, leading to the functional inactivation of the CYP27B1 gene. Clinically, all the patients required continued calcitriol treatment and the clinical presentations were consistent with the complete loss of vitamin D1α-hydroxylase activity. In conclusion, three novel mutations have been identified. All of them caused frameshift and truncated proteins. The silent c.1215 T>C SNV has no effect on pre-mRNA splicing and it is likely a novel SNP. The current study further expands the CYP27B1 mutation spectrum.

Hypophosphatemic rickets caused by a novel splice donor site mutation and activation of two cryptic splice donor sites in the PHEX gene.

X-linked hypophosphatemic rickets (XLH) is the most common inherited form of rickets. XLH is caused by inactivating mutations in the PHEX gene and is transmitted as an X-linked dominant disorder. We investigated PHEX mutation in a sporadic Turkish girl with hypophosphatemic rickets. The patient was 2 years of age with a complaint of inability to walk. She had bowing of legs and growth retardation. Laboratory data showed normal calcium, low phosphate with markedly elevated ALP, and low phosphate renal tubular reabsorption. She was treated with Calcitriol 0.5 mg/kg/day and oral phosphate supplement with good response. The entire coding region of PHEX gene was sequenced from patient's peripheral leukocyte DNA and a novel 13 bp deletion at the donor splice site of exon5 was found (c.663+12del). Instead of using the donor splice site of intron 4 to splice out exon 5 and intron 5, the spliceosome utilized two nearby cryptic donor splice sites (5' splice site) to splice out intron 4, resulting in two smaller transcripts. Both of them could not translate into functional proteins due to frameshift. Her parents did not carry the mutation, indicating that this is a de novo PHEX mutation likely resulting from mutagenesis of X chromosome in paternal germ cells. We conclude that c.663+12del is a novel mutation that can activate nearby cryptic 5' splice sites. The selection of cryptic 5' splice sites adds the complexity of cell's splicing mechanisms. The current study extends the database of PHEX mutation and cryptic 5' splice sites.