Macromolecular changes accompanying immortalization and tumorigenic conversion in a human fibroblast model system

Targeted deletion of the zebrafish actin-bundling protein L-plastin (lcp1)

Regulation of the cytoskeleton is essential for cell migration in health and disease. Lymphocyte cytosolic protein 1 (lcp1, also called L-plastin) is a hematopoietic-specific actin-bundling protein that is highly conserved in zebrafish, mice and humans. In addition, L-plastin expression is documented as both a genetic marker and a cellular mechanism contributing to the invasiveness of tumors and transformed cell lines. Despite L-plastin’s role in both immunity and cancer, in zebrafish there are no direct studies of its function, and no mutant, knockout or reporter lines available. Using CRISPR-Cas9 genome editing, we generated null alleles of zebrafish lcp1 and examined the phenotypes of these fish throughout the life cycle. Our editing strategy used gRNA to target the second exon of lcp1, producing F0 mosaic fish that were outcrossed to wild types to confirm germline transmission. F1 heterozygotes were then sequenced to identify three unique null alleles, here called ‘Charlie’, ‘Foxtrot’ and ‘Lima’. In silico, each allele truncates the endogenous protein to less than 5% normal size and removes both essential actin-binding domains (ABD1 and ABD2). Although none of the null lines express detectable LCP1 protein, homozygous mutant zebrafish (-/-) can develop and reproduce normally, a finding consistent with that of the L-plastin null mouse (LPL -/-). However, such mice do have a profound immune defect when challenged by lung bacteria. Interestingly, we observed reduced long-term survival of zebrafish lcp1 -/- homozygotes (~30% below the expected numbers) in all three of our knockout lines, with greatest mortality corresponding to the period (4–6 weeks post-fertilization) when the innate immune system is functional, but the adaptive immune system is not yet mature. This suggests that null zebrafish may have reduced capacity to combat opportunistic infections, which are more easily transmissible in the aquatic environment. Overall, our novel mutant lines establish a sound genetic model and an enhanced platform for further studies of L-plastin gene function in hematopoiesis and cancer.

Primary skin fibroblasts as a model of Parkinson's disease

Parkinson's disease is the second most frequent neurodegenerative disorder. While most cases occur sporadic mutations in a growing number of genes including Parkin (PARK2) and PINK1 (PARK6) have been associated with the disease. Different animal models and cell models like patient skin fibroblasts and recombinant cell lines can be used as model systems for Parkinson's disease. Skin fibroblasts present a system with defined mutations and the cumulative cellular damage of the patients. PINK1 and Parkin genes show relevant expression levels in human fibroblasts and since both genes participate in stress response pathways, we believe fibroblasts advantageous in order to assess, e.g. the effect of stressors. Furthermore, since a bioenergetic deficit underlies early stage Parkinson's disease, while atrophy underlies later stages, the use of primary cells seems preferable over the use of tumor cell lines. The new option to use fibroblast-derived induced pluripotent stem cells redifferentiated into dopaminergic neurons is an additional benefit. However, the use of fibroblast has also some drawbacks. We have investigated PARK6 fibroblasts and they mirror closely the respiratory alterations, the expression profiles, the mitochondrial dynamics pathology and the vulnerability to proteasomal stress that has been documented in other model systems. Fibroblasts from patients with PARK2, PARK6, idiopathic Parkinson's disease, Alzheimer's disease, and spinocerebellar ataxia type 2 demonstrated a distinct and unique mRNA expression pattern of key genes in neurodegeneration. Thus, primary skin fibroblasts are a useful Parkinson's disease model, able to serve as a complement to animal mutants, transformed cell lines and patient tissues.

F-actin and myosin II binding domains in supervillin

Detergent-resistant membranes contain signaling and integral membrane proteins that organize cholesterol-rich domains called lipid rafts. A subset of these detergent-resistant membranes (DRM-H) exhibits a higher buoyant density ( approximately 1.16 g/ml) because of association with membrane skeleton proteins, including actin, myosin II, myosin 1G, fodrin, and an actin- and membrane-binding protein called supervillin (Nebl, T., Pestonjamasp, K. N., Leszyk, J. D., Crowley, J. L., Oh, S. W., and Luna, E. J. (2002) J. Biol. Chem. 277, 43399-43409). To characterize interactions among DRM-H cytoskeletal proteins, we investigated the binding partners of the novel supervillin N terminus, specifically amino acids 1-830. We find that the supervillin N terminus binds directly to myosin II, as well as to F-actin. Three F-actin-binding sites were mapped to sequences within amino acids approximately 280-342, approximately 344-422, and approximately 700-830. Sequences with combinations of these sites promote F-actin cross-linking and/or bundling. Supervillin amino acids 1-174 specifically interact with the S2 domain in chicken gizzard myosin and nonmuscle myosin IIA (MYH-9) but exhibit little binding to skeletal muscle myosin II. Direct or indirect binding to filamin also was observed. Overexpression of supervillin amino acids 1-174 in COS7 cells disrupted the localization of myosin IIB without obviously affecting actin filaments. Taken together, these results suggest that supervillin may mediate actin and myosin II filament organization at cholesterol-rich membrane domains.

Regulation of microfilament organization and anchorage-independent growth by tropomyosin 1

Variants of chemically immortalized Syrian hamster embryo cells that had either retained (supB+) or lost (supB-) the ability to suppress tumorigenicity when hybridized with a fibrosarcoma cell line were subcloned. Both supB cell types are nontumorigenic; however, the supB- but not supB+ cells exhibit conditional anchorage-independent growth. Alterations of actin microfilament organization were observed in supB- but not supB+ cells that corresponded to a significant reduction of the actin-binding protein tropomyosin 1 (TM-1) in subB- cells. To examine the possibility of a direct relationship between TM-1 expression and the subB- phenotype, subB+ cells were transfected with an expression vector containing the TM-1 cDNA in an antisense orientation. The antisense-induced reduction of TM-1 levels in supB+ clones caused a microfilament reorganization and conferred anchorage-independent growth potential that were indistinguishable from those characteristic of supB- cells. These data provide direct evidence that TM-1 regulates both microfilament organization and anchorage-independent growth and suggest that microfilament alterations are sufficient for anchorage-independent growth.