C-terminal BRE overexpression in 11q23-rearranged and t(8;16) acute myeloid leukemia is caused by intragenic transcription initiation.

Overexpression of the BRE (brain and reproductive organ-expressed) gene defines a distinct pediatric and adult acute myeloid leukemia (AML) subgroup. Here we identify a promoter enriched for active chromatin marks in BRE intron 4 causing strong biallelic expression of a previously unknown C-terminal BRE transcript. This transcript starts with BRE intron 4 sequences spliced to exon 5 and downstream sequences, and if translated might code for an N terminally truncated BRE protein. Remarkably, the new BRE transcript was highly expressed in over 50% of 11q23/KMT2A (lysine methyl transferase 2A)-rearranged and t(8;16)/KAT6A-CREBBP cases, while it was virtually absent from other AML subsets and normal tissues. In gene reporter assays, the leukemia-specific fusion protein KMT2A-MLLT3 transactivated the intragenic BRE promoter. Further epigenome analyses revealed 97 additional intragenic promoter marks frequently bound by KMT2A in AML with C-terminal BRE expression. The corresponding genes may be part of a context-dependent KMT2A-MLLT3-driven oncogenic program, because they were higher expressed in this AML subtype compared with other groups. C-terminal BRE might be an important contributor to this program because in a case with relapsed AML, we observed an ins(11;2) fusing CHORDC1 to BRE at the region where intragenic transcription starts in KMT2A-rearranged and KAT6A-CREBBP AML.

Culture of impure human islet fractions in the presence of alpha-1 antitrypsin prevents insulin cleavage and improves islet recovery.

BACKGROUND: Exocrine tissue is commonly cotransplanted with islets in autografting and allotransplantation of impure preparations. Proteases and insulin are released by acinar cells and islets, respectively, during pretransplantation culture and also systemically after transplantation. We hypothesized that released proteases could cleave insulin molecules and that addition of alpha-1 antitrypsin (A1AT) to impure islet cultures would block this cleavage, improving islet recovery and function. METHODS: Trypsin, chymotrypsin, and elastase (TCE) activity and insulin levels were measured in culture supernates of pure (n = 5) and impure (n = 5) islet fractions, which were isolated from deceased donors. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was used to detect insulin after incubation with proteases. We assessed the effects of A1AT supplementation (0.5 mg/mL; n = 4] on TCE activity, insulin levels, culture recovery, and islet quality. The ultrastructure of islets exposed to TCE versus control medium was examined using electron microscopy (EM). RESULTS: Protease (TCE) activity in culture supernatants was indirectly proportional to the percentage purity of islets: pure, impure, or highly impure. Increasingly lower levels of insulin were detected in culture supernatants when higher protease activity levels were present. Insulin levels measured from supernatants of impure and highly impure islet preparations were 61 +/- 23.7% and 34 +/- 33% of that in pure preparations, respectively. Incubation with commercially available proteases (TCE) or exocrine acinar cell supernatant cleaved insulin molecules as assessed using SDS-PAGE. Addition of A1AT to impure islet preparations reduced protease activity and restored normal insulin levels as detected using enzyme-linked immunosorbent assay (ELISA) and SDS-PAGE of culture supernates. A1AT improved insulin levels to 98% +/- 1.3% in impure and 78% +/- 34.2% in highly impure fractions compared with pure islet fractions. A1AT supplementation improved postculture recovery of islets in impure preparations compared with nontreated controls (72% +/- 9% vs 47% +/- 15%). Islet viability as measured using membrane integrity assays was similar in both the control (98% +/- 2%) and the A1AT-treated groups (99% +/- 1%). EM results revealed a reduction or absence of secretory granules after exposure to proteases (TCE). CONCLUSION: Culture of impure human islet fractions in the presence of A1AT prevented insulin cleavage and improved islet recovery. A1AT supplementation of islet culture media, therefore, may increase the proportion of human islet products that meet release criteria for transplantation.

Collagen IV-dependent ERK activation in human Caco-2 intestinal epithelial cells requires focal adhesion kinase.

Integrin-initiated extracellular signal-regulated kinase (ERK) activation by matrix adhesion may require focal adhesion kinase (FAK) or be FAK-independent via caveolin and Shc. This remains controversial for fibroblast and endothelial cell adhesion to fibronectin and is less understood for other matrix proteins and cells. We investigated Caco-2 intestinal epithelial cell ERK activation by collagen I and IV, laminin, and fibronectin. Collagens or laminin, but not fibronectin, stimulated tyrosine phosphorylation of FAK, paxillin, and p130(cas) and activated ERK1/2. Shc, tyrosine-phosphorylated by matrix adhesion in many cells, was not phosphorylated in Caco-2 cells in response to any matrix. Caveolin expression did not affect Caco-2 Shc phosphorylation in response to fibronectin. FAK, ERK, and p130(cas) tyrosine phosphorylation were activated after 10-min adhesion to collagen IV. FAK activity increased for 45 min after collagen IV adhesion and persisted for 2 h, while p130(cas) phosphorylation increased only slightly after 10 min. ERK activity peaked at 10 min, declined after 30 min, and returned to base line after 1 h. Transfection with FAK-related nonkinase, but not substrate domain deleted p130(cas), strongly inhibited ERK2 activation in response to collagen IV, indicating Caco-2 ERK activation is at least partly regulated by FAK.

Adenovirus-mediated gene transfer in primary murine adipocytes.

The transfer of genes into primary murine adipocytes using an adenovirus system has been developed. A recombinant adenovirus was constructed (expressing green fluorescent protein [GFP] under the control of the strong cytomegalovirus [CMV] promoter and a luciferase reporter gene under the control of the weak adipocyte promoter keratinocyte lipid-binding protein [KLBP/FABP5]) and incubated with primary adipocytes from C57BL/6J mice. Analysis of infected cells by confocal microscopy detected GFP expression in both the cytoplasm and nucleus of adipocytes with a 64% efficiency of infection. To demonstrate the applicability of this method in the study of gene regulation, adenovirus-infected adipocytes exhibited significant levels of luciferase activity even from a weak promoter. TPA treatment of infected adipocytes increased luciferase activity, consistent with previous studies indicating that the KLBP/FABP5 gene is up-regulated by phorbol esters. These results provide an efficient, convenient, and sensitive method to transiently infect primary murine adipocytes, facilitating protein expression or permitting analysis of reporter gene activity from both viral and endogenous promoters.

Human caco-2 motility redistributes FAK and paxillin and activates p38 MAPK in a matrix-dependent manner.

The signals involved in restitution during mucosal healing are poorly understood. We compared focal adhesion kinase (FAK) and paxillin protein and phosphorylation, extracellular signal-regulated kinase (ERK) 1, ERK2, and p38 activation, as well as FAK and paxillin organization in static and migrating human intestinal Caco-2 cells on matrix proteins and anionically derivatized polystyrene dishes (tissue culture plastic). We also studied effects of FAK, ERK, and p38 blockade in a monolayer-wounding model. Compared with static cells, cells migrating across matrix proteins matrix-dependently decreased membrane/cytoskeletal FAK and paxillin and cytosolic FAK. Tyrosine phosphorylated FAK and paxillin changed proportionately to FAK and paxillin protein. Conversely, cells migrating on plastic increased FAK and paxillin protein and phosphorylation. Migration matrix-dependently activated p38 and inactivated ERK1 and ERK2. Total p38, ERK1, and ERK2 did not change. Caco-2 motility was inhibited by transfection of FRNK (the COOH-terminal region of FAK) and PD-98059, a mitogen-activated protein kinase-ERK kinase inhibitor, but not by SB-203580, a p38 inhibitor, suggesting that FAK and ERK modulate Caco-2 migration. In contrast to adhesion-induced phosphorylation, matrix may regulate motile intestinal epithelial cells by altering amounts and distribution of focal adhesion plaque proteins available for phosphorylation as well as by p38 activation and ERK inactivation. Motility across plastic differs from migration across matrix.

Alpha integrin subunits regulate human (Caco-2) intestinal epithelial proliferation and phenotype.

BACKGROUND/AIMS: Cell-matrix interactions influence intestinal epithelial biology, but the whether specific integrin heterodimers exert different effects is unclear. METHODS: We used functional antibodies to investigate effects of the alpha2, alpha3, alpha5, and alpha6 integrin subunits on proliferation and differentiation of human intestinal Caco-2 cells on laminin. Cells seeded onto laminin-coated inserts in defibronectinized medium were treated with functional antibodies or normal IgG for 72 hrs and proliferation, alkaline phosphatase and dipeptidyl dipeptidase specific activity were measured. RESULTS: Caco-2 adhesion to antibody to each alpha-integrin subunit stimulated tyrosine phosphorylation of Focal Adhesion Kinase and paxillin, suggesting that these antibodies intiate integrin-related tyrosine signalling. Proliferation was inhibited by anti-alpha2 and anti-alpha3, but stimulated by anti-alpha6. Alkaline phosphatase and dipeptidyl dipeptidase specific activity were promoted by alpha2 blockade but decreased after a6 blockade. Proliferative blockade using mitomycin C or hydroxyurea prevented the effects of alpha2 ligation on differentiation, but the decrease in alkaline phosphatase specific activity observed after a6 integrin subunit blockade was preserved even after proliferative blockade. CONCLUSION: lntegrin heterodimers modulate human Caco-2 intestinal epithelial biology on laminin in an alpha-subunit specific manner. The different effects of anti-alpha2 and anti-alpha6 may reflect competitition by these antibodies with integrin subunit interactions with laminin as well as initiation of their own signals or different functions for the alpha6 integrin subunit

Effects of increased ambient pressure on colon cancer cell adhesion.

Forces such as strain modulate intestinal epithelial biology. Shear and pressure influence other cells. The effects of pressure on human colon cancer cells are poorly understood. Increasing ambient pressure for 30 min by 15 mm Hg over atmospheric stimulated adhesion to matrix proteins of four human colon cancer cell lines and primary cells from three human colon cancers, but not bovine aortic smooth-muscle cells. This effect was energy dependent and cation dependent (blocked by azide and chelation), accompanied by tyrosine phosphorylation of intracellular proteins including focal adhesion kinase, and blocked by tyrosine kinase inhibition (genistein, tyrphostin, and erbstatin) and a functional antibody to the beta1 integrin subunit. Although pressure stimulated adhesion even in a balanced salt solution, baseline and pressure-stimulated adhesion were each substantially diminished in the absence of serum. These data suggest that relatively low levels of increased pressure may stimulate malignant colonocyte adhesion by a cation-dependent beta1-integrin-mediated mechanism, perhaps via focal adhesion kinase-related tyrosine phosphorylation. In addition to elucidating another aspect of physical force regulation of colonocyte biology, these findings may be relevant to the effects of increased pressure engendered by colonic peristalsis, surgical manipulation, or laparoscopic surgery on colon cancer cell adhesion.

William Cheselden: anatomist, surgeon, and medical illustrator.

William Cheselden was Great Britain's foremost surgeon/scientist in the first half of the 18th century. Cheselden directly challenged the Company of Barber-Surgeons' exclusive right to control dissection in London by being the first to conduct a regular series of anatomy lectures and demonstrations outside of the Company's Hall. He incorporated his lecture syllabus into a handbook of anatomy, The Anatomy of the Humane Body, which was used by students for nearly 100 years. Cheselden also wrote the text and drew the illustrations for a majestic atlas of comparative osteology, the Osteographia, or the Anatomy of the Bones. Cheselden used his superior knowledge of anatomy to reduce the morbidity and mortality associated with perineal lithotomy, one of the few operations possible in his era. Sagacious and pragmatic, Cheselden recognized that the enlightened practice of surgery beginning to take root in 18th-century London could flourish only under an autonomous body of surgeons. Cheselden used his personal funds and political skills to urge Parliament to pass legislation for the dissolution of the combined Company of Barber-Surgeons and the establishment of separate and distinct Surgeons' and Barbers' Companies. After disjoinder of the two groups on May 2, 1745, Cheselden served as one of the Wardens of the new Company of Surgeons--a predecessor of the Royal College of Surgeons of England. In 1746, Cheselden, who helped design the first Surgeons' Hall, served as the Company's Master.

Loss of matrix-dependent cytoskeletal tyrosine kinase signals may regulate intestinal epithelial differentiation during mucosal healing.

Intestinal epithelial restitution and the migratory phenotype appear regulated by the extracellular matrix. Since integrin-associated adhesion to matrix triggers tyrosine kinase activity, we hypothesized that matrix-specific tyrosine kinase signals might modulate the intestinal epithelial migratory phenotype, particularly via focal adhesion kinase. Caco-2 cells were seeded at two densities on collagen I, laminin, fibronectin, and tissue culture plastic. Four days later the first cells were confluent, whereas the second cells were not contact inhibited and expressed migratory lamellipodia. Cells were fractionated into membrane/cytoskeletal and cytosolic fractions. Cytoskeletal tyrosine kinase activity in static cells was matrix dependent and, unlike cytoscolic tyrosine kinase, correlated with adhesion, highest on collagen and lowest on plastic. Migrating cells exhibited matrix-dependent increases in cystosolic tyrosine kinase activity. Cytosolic changes in tyrosine kinase activity in motile cells exceeded membrane/cytoskeletal changes. However, matrix-dependent variations in increase in cytosolic tyrosine kinase activity correlated inversely with changes in cytoskeletal tyrosine kinase activity, suggesting cytoskeletal tyrosine kinase translocation to the cytosol during motility. Indeed cytoskeletal focal adhesion kinase activity decreased during migration on collagen. Tyrosine kinase inhibition by genistein both inhibited migration and stimulated expression of brush-border enzymes downregulated during motility. Although enterocyte-matrix interactions alter both cytosolic and cytoskeletal tyrosine kinase activity, matrix-dependent cytoskeletal events are likely to regulate adhesion and differentiation in static cells. Loss of matrix-dependent cytoskeletal tyrosine kinase signals such as focal adhesion kinase during restitution may trigger a phenotypic switch to the "dedifferentiated" migrating intestinal epithelial phenotype.

Cytoplasmic dynein is required for the nuclear attachment and migration of centrosomes during mitosis in Drosophila.

Cytoplasmic dynein is a multisubunit minus-end-directed microtubule motor that serves multiple cellular functions. Genetic studies in Drosophila and mouse have demonstrated that dynein function is essential in metazoan organisms. However, whether the essential function of dynein reflects a mitotic requirement, and what specific mitotic tasks require dynein remains controversial. Drosophila is an excellent genetic system in which to analyze dynein function in mitosis, providing excellent cytology in embryonic and somatic cells. We have used previously characterized recessive lethal mutations in the dynein heavy chain gene, Dhc64C, to reveal the contributions of the dynein motor to mitotic centrosome behavior in the syncytial embryo. Embryos lacking wild-type cytoplasmic dynein heavy chain were analyzed by in vivo analysis of rhodamine-labeled microtubules, as well as by immunofluorescence in situ methods. Comparisons between wild-type and Dhc64C mutant embryos reveal that dynein function is required for the attachment and migration of centrosomes along the nuclear envelope during interphase/prophase, and to maintain the attachment of centrosomes to mitotic spindle poles. The disruption of these centrosome attachments in mutant embryos reveals a critical role for dynein function and centrosome positioning in the spatial organization of the syncytial cytoplasm of the developing embryo.

Human Caco-2 intestinal epithelial motility is associated with tyrosine kinase and cytoskeletal focal adhesion kinase signals.

Intestinal epithelial cells assume a specialized phenotype adapted to motility and mucosal healing during mucosal restitution. Since cell-matrix interactions initiate tyrosine kinase (TK) signals, we hypothesized that the regulation of the intestinal epithelial migratory phenotype may also involve TK signals, particularly via focal adhesion kinase (FAK). Caco-2 cells were seeded simultaneously at 26,000 and 6000 cells/cm2. After 4 days, the first cells were confluent, while cells in the second population were not contact-inhibited and expressed migrating lamellipodia. Cells were fractionated into Triton X-100-soluble (membrane/cytoskeletal) and -insoluble (cytosolic) fractions. TK activity in each fraction was assayed by ELISA using a synthetic substrate. FAK protein was assessed by immunoprecipitation with monoclonal anti-FAK and Western blotting. Because active FAK autophosphorylates, we also measured FAK tyrosine phosphorylation, immunoprecipitating with anti-FAK and then Western blotting for phosphotyrosine. TK activity was increased in both cytosolic and membrane/cytoskeletal fractions of migrating cells by 17.6 +/- 3.6 and 28.9 +/- 4.1%, respectively, compared to static cells (n = 11, P < 0.01). FAK protein increased in the cytosolic fraction by 90.4 +/- 20.0% (n = 5, P = 0.01), but did not change in the membrane/cytoskeletal fraction. Tyrosine phosphorylated FAK increased by 62.8 +/- 21.4% in the cytosolic fraction of migrating cells but also by 46.6 +/- 38.4% in the membrane/cytoskeletal fraction (n = 5, P < 0.05). These data suggest that intestinal epithelial cell migration is associated with increases in both cytosolic and cytoskeletal TK activity and upregulation of cytosolic FAK protein. The increase in cytoskeletal FAK phosphorylation without increased FAK protein suggests autophosphorylation and increased cytoskeletal FAK activity. The migrating intestinal epithelial phenotype may therefore be modulated by TK signals including cytoskeletal FAK activity.

RLF2, a subunit of yeast chromatin assembly factor-I, is required for telomeric chromatin function in vivo.

In the yeast Saccharomyces cerevisiae, telomere repeat DNA is assembled into a specialized heterochromatin-like complex that silences the transcription of adjacent genes. The general DNA-binding protein Rap1p binds telomere DNA repeats, contributes to telomere length control and to telomeric silencing, and is a major component of telomeric chromatin. We identified Rap1p localization factor 2 (RLF2) in a screen for genes that alleviate antagonism between telomere and centromere sequences on plasmids. In rlf2 mutants, telomeric chromatin is perturbed: Telomeric silencing is reduced and Rap1p localization is altered. In wild-type cells, Rap1p and telomeres localize to bright perinuclear foci. In rlf2 strains, the number of Rap1p foci is increased, Rap1p staining is more diffuse throughout the nucleus, Rap1p foci are distributed in a much broader perinuclear domain, and nuclear volume is 50% larger. Despite the altered distribution of Rap1p in rlf2 mutant cells, fluorescence in situ hybridization to subtelomeric repeats shows that the distribution of telomeric DNA is similar in wild-type and mutant cells. Thus in rlf2 mutant cells, the distribution of Rap1p does not reflect the distribution of telomeric DNA. RLF2 encodes a highly charged coiled-coil protein that has significant similarity to the p150 subunit of human chromatin assembly factor-I(hCAF-I), a complex that is required for the DNA replication-dependent assembly of nucleosomes from newly synthesized histones in vitro. Furthermore, RLF2 is identical to CAC1, a subunit of yeast chromatin assembly factor-I (yCAF-I) which assembles nucleosomes in vitro. In wild-type cells, epitope-tagged Rlf2p expressed from the GAL10 promoter localizes to the nucleus with a pattern distinct from that of Rap1p, suggesting that Rlf2p is not a component of telomeric chromatin. This study provides evidence that yCAF-I is required for the function and organization of telomeric chromatin in vivo. We propose that Rlf2p facilitates the efficient and timely assembly of histones into telomeric chromatin.

Desensitization of the neurokinin 1 receptor is mediated by the receptor carboxy-terminal region, but is not caused by receptor internalization.

The carboxy-terminal cytoplasmic regions of the rat neurokinin 1 (substance P) and neurokinin 2 (neurokinin A) receptors have been exchanged to determine if this region of the neurokinin 1 receptor is involved in its desensitization. When expressed at similar levels in stably transfected Chinese hamster ovary (CHO) cell lines, receptors containing the carboxy-terminal region of the neurokinin 1 receptor desensitized significantly more (as measured by reduction of the inositol 1,4,5-trisphosphate response) when preexposed for 1 min to 1 microM neurokinin, indicating a role for the carboxy-terminal region of the neurokinin 1 receptor in its desensitization. Measurement of receptor internalization using radiolabeled neurokinins (0.3 nM) indicated that approximately 75-80% of the receptors were internalized in each cell line after 10 min at 37 degrees C, with no observable correlation between neurokinin receptor desensitization and internalization. Measurement of loss of receptor surface sites for cell lines CHO NK1 and CHO NK1NK2 following exposure to 1 microM substance P also indicated no obvious relationship between the percent desensitization and percent of receptors internalized. Also, two inhibitors of neurokinin 1 receptor internalization, phenylarsine oxide and hyperosmolar sucrose, did not inhibit neurokinin 1 receptor desensitization. The protein kinase inhibitors Ro 31-8220, staurosporine, and Zn2+ had no effect on neurokinin 1 receptor desensitization, indicating that the kinases affected by these agents are not rate-limiting in neurokinin 1 receptor desensitization in this system.

Centrin plays an essential role in microtubule severing during flagellar excision in Chlamydomonas reinhardtii.

Previously, we reported that flagellar excision in Chlamydomonas reinhardtii is mediated by an active process whereby microtubules are severed at select sites within the flagellar-basal body transition zone (Sanders, M. A., and J. L. Salisbury. 1989. J. Cell Biol. 108:1751-1760). At the time of flagellar excision, stellate fibers of the transition zone contract and displace the microtubule doublets of the axoneme inward. The resulting shear force and torsional load generated during inward displacement leads to microtubule severing immediately distal to the central cylinder of the transition zone. In this study, we have used a detergent-extracted cell model of Chlamydomonas that allows direct experimental access to the molecular machinery responsible for microtubule severing without the impediment of the plasma membrane. We present four independent lines of experimental evidence for the essential involvement of centrin-based stellate fibers of the transition zone in the process of flagellar excision: (a) Detergent-extracted cell models excise their flagella in response to elevated, yet physiological, levels of free calcium. (b) Extraction of cell models with buffers containing the divalent cation chelator EDTA leads to the disassembly of centrin-based fibers and to the disruption of transition zone stellate fiber structure. This treatment results in a complete loss of flagellar excision competence. (c) Three separate anti-centrin monoclonal antibody preparations, which localize to the stellate fibers of the transition zone, specifically inhibit contraction of the stellate fibers and block calcium-induced flagellar excision, while control antibodies have no inhibitory effect. Finally, (d) cells of the centrin mutant vfl-2 (Taillon, B., S. Adler, J. Suhan, and J. Jarvik. 1992. J. Cell Biol. 119:1613-1624) fail to actively excise their flagella following pH shock in living cells or calcium treatment of detergent-extracted cell models. Taken together, these observations demonstrate that centrin-based fiber contraction plays a fundamental role in microtubule severing at the time of flagellar excision in Chlamydomonas.

Cloning of a cDNA encoding human centrin, an EF-hand protein of centrosomes and mitotic spindle poles.

A human cDNA expression library was screened using anti-centrin antibodies to obtain a cDNA clone encoding human centrin. The cDNA clone contains an open reading frame of 516 base pairs and predicts a product of 172 amino acids with a calculated molecular mass of 19,528 and a pI of 4.61. Sequence analysis demonstrates that human centrin and centrins from higher plants, protozoa, algae, Xenopus and the yeast CDC31 gene product are closely related members of a subfamily of the EF-hand superfamily of calcium-binding proteins. The human centrin sequence has four putative calcium-binding domains as defined by the EF-hand consensus. Immunoprecipitation and western blot studies from HeLa cells confirm that human centrin is a protein of approximately 20,000 M(r) as predicted from the cDNA clone. Indirect immunofluorescence analysis of HeLa cells demonstrates that centrin is localized at the centrosome of interphase cells and that it redistributes to the region of the spindle poles during mitosis. When taken together with earlier genetic studies, these results demonstrate that centrin is a ubiquitous component of centrosomes and mitotic spindle poles of diverse organisms and suggest that centrin plays a role in centrosome separation at the time of mitosis.

Phosphorylation of nuclear and flagellar basal apparatus proteins during flagellar regeneration in Chlamydomonas reinhardtii.

The antiphosphoprotein monoclonal antibody MPM-2 was used to investigate protein phosphorylation during flagellar regeneration in Chlamydomonas reinhardtii. MPM-2 recognizes a phosphorylated epitope and detects several Chlamydomonas proteins by Western immunoblot analysis. Two MPM-2 reactive proteins (34 and 90 kD) increase in Western immunoblot intensity after flagellar excision and decrease in intensity during flagellar regeneration. Immunofluorescence and immunogold labeling revealed MPM-2 staining within the nucleus, especially towards the nuclear periphery, the flagellar basal apparatus, and the nucleus-basal body connector after flagellar excision. Comparison of MPM-2 reactivity in wild-type cells and in the mutant bald-2, which lacks functional basal bodies, demonstrates that the 34-kD protein is localized in the nucleus and the 90-kD protein is localized in the flagellar basal region. MPM-2 reactivity is observed in cells competent for flagellar regeneration. However, when cells were treated with the kinase inhibitor, staurosporine, MPM-2 reactivity did not increase after flagellar excision and flagellar regeneration was impaired. These observations suggest that phosphorylation of the 34- and 90-kD proteins may be important for flagellar regrowth. Possible roles for phosphorylation in flagellar regeneration include transcriptional activation and transport of flagellar precursors to the base of the growing flagella.