Mammalian UPF3A and UPF3B can activate nonsense-mediated mRNA decay independently of their exon junction complex binding.

Nonsense-mediated mRNA decay (NMD) is governed by the three conserved factors-UPF1, UPF2, and UPF3. While all three are required for NMD in yeast, UPF3B is dispensable for NMD in mammals, and its paralog UPF3A is suggested to only weakly activate or even repress NMD due to its weaker binding to the exon junction complex (EJC). Here, we characterize the UPF3A/B-dependence of NMD in human cell lines deleted of one or both UPF3 paralogs. We show that in human colorectal cancer HCT116 cells, NMD can operate in a UPF3B-dependent and -independent manner. While UPF3A is almost dispensable for NMD in wild-type cells, it strongly activates NMD in cells lacking UPF3B. Notably, NMD remains partially active in cells lacking both UPF3 paralogs. Complementation studies in these cells show that EJC-binding domain of UPF3 paralogs is dispensable for NMD. Instead, the conserved "mid" domain of UPF3 paralogs is consequential for their NMD activity. Altogether, our results demonstrate that the mammalian UPF3 proteins play a more active role in NMD than simply bridging the EJC and the UPF complex.

Analysis of programmed frameshifting during translation of prfB in Flavobacterium johnsoniae.

Ribosomes of Bacteroidia fail to recognize Shine-Dalgarno (SD) sequences due to sequestration of the 3' tail of the 16S rRNA on the 30S platform. Yet in these organisms, the prfB gene typically contains the programmed +1 frameshift site with its characteristic SD sequence. Here, we investigate prfB autoregulation in Flavobacterium johnsoniae, a member of the Bacteroidia. We find that the efficiency of prfB frameshifting in F. johnsoniae is low (∼7%) relative to that in Escherichia coli (∼50%). Mutation or truncation of bS21 in F. johnsoniae increases frameshifting substantially, suggesting that anti-SD (ASD) sequestration is responsible for the reduced efficiency. The frameshift site of certain Flavobacteriales, such as Winogradskyella psychrotolerans, has no SD. In F. johnsoniae, this W. psychrotolerans sequence supports frameshifting as well as the native sequence, and mutation of bS21 causes no enhancement. These data suggest that prfB frameshifting normally occurs without SD-ASD pairing, at least under optimal laboratory growth conditions. Chromosomal mutations that remove the frameshift or ablate the SD confer subtle growth defects in the presence of paraquat or streptomycin, respectively, indicating that both the autoregulatory mechanism and the SD element contribute to F. johnsoniae cell fitness. Analysis of prfB frameshift sites across 2686 representative bacteria shows loss of the SD sequence in many clades, with no obvious relationship to genome-wide SD usage. These data reveal unexpected variation in the mechanism of frameshifting and identify another group of organisms, the Verrucomicrobiales, that globally lack SD sequences.

Identification of leader-trailer helices of precursor ribosomal RNA in all phyla of bacteria and archaea.

Ribosomal RNAs are transcribed as part of larger precursor molecules. In Escherichia coli, complementary RNA segments flank each rRNA and form long leader-trailer (LT) helices, which are crucial for subunit biogenesis in the cell. A previous study of 15 representative species suggested that most but not all prokaryotes contain LT helices. Here, we use a combination of in silico folding and covariation methods to identify and characterize LT helices in 4,464 bacterial and 260 archaeal organisms. Our results suggest that LT helices are present in all phyla, including Deinococcota, which had previously been suspected to lack LT helices. In very few organisms, our pipeline failed to detect LT helices for both 16S and 23S rRNA. However, a closer case-by-case look revealed that LT helices are indeed present but escaped initial detection. 3,618 secondary structure models, many well-supported by nucleotide covariation, were generated. These structures show a high degree of diversity. Yet, all exhibit extensive base-pairing between the leader and trailer strands, in line with a common and essential function.

Roles of the leader-trailer helix and antitermination complex in biogenesis of the 30S ribosomal subunit.

Ribosome biogenesis occurs co-transcriptionally and entails rRNA folding, ribosomal protein binding, rRNA processing, and rRNA modification. In most bacteria, the 16S, 23S and 5S rRNAs are co-transcribed, often with one or more tRNAs. Transcription involves a modified RNA polymerase, called the antitermination complex, which forms in response to cis-acting elements (boxB, boxA and boxC) in the nascent pre-rRNA. Sequences flanking the rRNAs are complementary and form long helices known as leader-trailer helices. Here, we employed an orthogonal translation system to interrogate the functional roles of these RNA elements in 30S subunit biogenesis in Escherichia coli. Mutations that disrupt the leader-trailer helix caused complete loss of translation activity, indicating that this helix is absolutely essential for active subunit formation in the cell. Mutations of boxA also reduced translation activity, but by only 2- to 3-fold, suggesting a smaller role for the antitermination complex. Similarly modest drops in activity were seen upon deletion of either or both of two leader helices, termed here hA and hB. Interestingly, subunits formed in the absence of these leader features exhibited defects in translational fidelity. These data suggest that the antitermination complex and precursor RNA elements help to ensure quality control during ribosome biogenesis.

Ribosomes lacking bS21 gain function to regulate protein synthesis in Flavobacterium johnsoniae.

Ribosomes of Bacteroidia (formerly Bacteroidetes) fail to recognize Shine-Dalgarno (SD) sequences even though they harbor the anti-SD (ASD) of 16S rRNA. Inhibition of SD-ASD pairing is due to sequestration of the 3' tail of 16S rRNA in a pocket formed by bS21, bS18, and bS6 on the 30S platform. Interestingly, in many Flavobacteriales, the gene encoding bS21, rpsU, contains an extended SD sequence. In this work, we present genetic and biochemical evidence that bS21 synthesis in Flavobacterium johnsoniae is autoregulated via a subpopulation of ribosomes that specifically lack bS21. Mutation or depletion of bS21 in the cell increases translation of reporters with strong SD sequences, such as rpsU'-gfp, but has no effect on other reporters. Purified ribosomes lacking bS21 (or its C-terminal region) exhibit higher rates of initiation on rpsU mRNA and lower rates of initiation on other (SD-less) mRNAs than control ribosomes. The mechanism of autoregulation depends on extensive pairing between mRNA and 16S rRNA, and exceptionally strong SD sequences, with predicted pairing free energies of < -13 kcal/mol, are characteristic of rpsU across the Bacteroidota. This work uncovers a clear example of specialized ribosomes in bacteria.

PP2A is a therapeutically targetable driver of cell fate decisions via a c-Myc/p21 axis in human and murine acute myeloid leukemia.

Dysregulated cellular differentiation is a hallmark of acute leukemogenesis. Phosphatases are widely suppressed in cancers but have not been traditionally associated with differentiation. In this study, we found that the silencing of protein phosphatase 2A (PP2A) directly blocks differentiation in acute myeloid leukemia (AML). Gene expression and mass cytometric profiling revealed that PP2A activation modulates cell cycle and transcriptional regulators that program terminal myeloid differentiation. Using a novel pharmacological agent, OSU-2S, in parallel with genetic approaches, we discovered that PP2A enforced c-Myc and p21 dependent terminal differentiation, proliferation arrest, and apoptosis in AML. Finally, we demonstrated that PP2A activation decreased leukemia-initiating stem cells, increased leukemic blast maturation, and improved overall survival in murine Tet2-/-Flt3ITD/WT and human cell-line derived xenograft AML models in vivo. Our findings identify the PP2A/c-Myc/p21 axis as a critical regulator of the differentiation/proliferation switch in AML that can be therapeutically targeted in malignancies with dysregulated maturation fate.

Targeting OCT3 attenuates doxorubicin-induced cardiac injury.

Doxorubicin is a commonly used anticancer agent that can cause debilitating and irreversible cardiac injury. The initiating mechanisms contributing to this side effect remain unknown, and current preventative strategies offer only modest protection. Using stem-cell-derived cardiomyocytes from patients receiving doxorubicin, we probed the transcriptomic landscape of solute carriers and identified organic cation transporter 3 (OCT3) (SLC22A3) as a critical transporter regulating the cardiac accumulation of doxorubicin. Functional validation studies in heterologous overexpression models confirmed that doxorubicin is transported into cardiomyocytes by OCT3 and that deficiency of OCT3 protected mice from acute and chronic doxorubicin-related changes in cardiovascular function and genetic pathways associated with cardiac damage. To provide proof-of-principle and demonstrate translational relevance of this transport mechanism, we identified several pharmacological inhibitors of OCT3, including nilotinib, and found that pharmacological targeting of OCT3 can also preserve cardiovascular function following treatment with doxorubicin without affecting its plasma levels or antitumor effects in multiple models of leukemia and breast cancer. Finally, we identified a previously unrecognized, OCT3-dependent pathway of doxorubicin-induced cardiotoxicity that results in a downstream signaling cascade involving the calcium-binding proteins S100A8 and S100A9. These collective findings not only shed light on the etiology of doxorubicin-induced cardiotoxicity, but also are of potential translational relevance and provide a rationale for the implementation of a targeted intervention strategy to prevent this debilitating side effect.

dsRBPBind: modeling the effect of RNA secondary structure on double-stranded RNA-protein binding.

MOTIVATION: RNA-binding proteins are fundamental to many cellular processes. Double-stranded RNA-binding proteins (dsRBPs) in particular are crucial for RNA interference, mRNA elongation, A-to-I editing, host defense, splicing and a multitude of other important mechanisms. Since dsRBPs require double-stranded RNA to bind, their binding affinity depends on the competition among all possible secondary structures of the target RNA molecule. Here, we introduce a quantitative model that allows calculation of the effective affinity of dsRBPs to any RNA given a principal affinity and the sequence of the RNA, while fully taking into account the entire secondary structure ensemble of the RNA. RESULTS: We implement our model within the ViennaRNA folding package while maintaining its O(N3) time complexity. We validate our quantitative model by comparing with experimentally determined binding affinities and stoichiometries for transactivation response element RNA-binding protein (TRBP). We also find that the change in dsRBP binding affinity purely due to the presence of alternative RNA structures can be many orders of magnitude and that the predicted affinity of TRBP for pre-miRNA-like constructs correlates with experimentally measured processing rates. AVAILABILITY AND IMPLEMENTATION: Our modified version of the ViennaRNA package is available for download at http://bioserv.mps.ohio-state.edu/dsRBPBind, is free to use for research and educational purposes, and utilizes simple get/set methods for footprint size, concentration, cooperativity, principal dissociation constant and overlap.

A nanoscale DNA force spectrometer capable of applying tension and compression on biomolecules.

Single molecule force spectroscopy is a powerful approach to probe the structure, conformational changes, and kinetic properties of biological and synthetic macromolecules. However, common approaches to apply forces to biomolecules require expensive and cumbersome equipment and relatively large probes such as beads or cantilevers, which limits their use for many environments and makes integrating with other methods challenging. Furthermore, existing methods have key limitations such as an inability to apply compressive forces on single molecules. We report a nanoscale DNA force spectrometer (nDFS), which is based on a DNA origami hinge with tunable mechanical and dynamic properties. The angular free energy landscape of the nDFS can be engineered across a wide range through substitution of less than 5% of the strand components. We further incorporate a removable strut that enables reversible toggling of the nDFS between open and closed states to allow for actuated application of tensile and compressive forces. We demonstrate the ability to apply compressive forces by inducing a large bend in a 249bp DNA molecule, and tensile forces by inducing DNA unwrapping of a nucleosome sample. These results establish a versatile tool for force spectroscopy and robust methods for designing nanoscale mechanical devices with tunable force application.

Structural basis of sequestration of the anti-Shine-Dalgarno sequence in the Bacteroidetes ribosome.

Genomic studies have indicated that certain bacterial lineages such as the Bacteroidetes lack Shine-Dalgarno (SD) sequences, and yet with few exceptions ribosomes of these organisms carry the canonical anti-SD (ASD) sequence. Here, we show that ribosomes purified from Flavobacterium johnsoniae, a representative of the Bacteroidetes, fail to recognize the SD sequence of mRNA in vitro. A cryo-electron microscopy structure of the complete 70S ribosome from F. johnsoniae at 2.8 Å resolution reveals that the ASD is sequestered by ribosomal proteins bS21, bS18 and bS6, explaining the basis of ASD inhibition. The structure also uncovers a novel ribosomal protein-bL38. Remarkably, in F. johnsoniae and many other Flavobacteriia, the gene encoding bS21 contains a strong SD, unlike virtually all other genes. A subset of Flavobacteriia have an alternative ASD, and in these organisms the fully complementary sequence lies upstream of the bS21 gene, indicative of natural covariation. In other Bacteroidetes classes, strong SDs are frequently found upstream of the genes for bS21 and/or bS18. We propose that these SDs are used as regulatory elements, enabling bS21 and bS18 to translationally control their own production.

Zebrafish rbm8a and magoh mutants reveal EJC developmental functions and new 3'UTR intron-containing NMD targets.

Many post-transcriptional mechanisms operate via mRNA 3'UTRs to regulate protein expression, and such controls are crucial for development. We show that homozygous mutations in two zebrafish exon junction complex (EJC) core genes rbm8a and magoh leads to muscle disorganization, neural cell death, and motor neuron outgrowth defects, as well as dysregulation of mRNAs subjected to nonsense-mediated mRNA decay (NMD) due to translation termination ≥ 50 nts upstream of the last exon-exon junction. Intriguingly, we find that EJC-dependent NMD also regulates a subset of transcripts that contain 3'UTR introns (3'UI) < 50 nts downstream of a stop codon. Some transcripts containing such stop codon-proximal 3'UI are also NMD-sensitive in cultured human cells and mouse embryonic stem cells. We identify 167 genes that contain a conserved proximal 3'UI in zebrafish, mouse and humans. foxo3b is one such proximal 3'UI-containing gene that is upregulated in zebrafish EJC mutant embryos, at both mRNA and protein levels, and loss of foxo3b function in EJC mutant embryos significantly rescues motor axon growth defects. These data are consistent with EJC-dependent NMD regulating foxo3b mRNA to control protein expression during zebrafish development. Our work shows that the EJC is critical for normal zebrafish development and suggests that proximal 3'UIs may serve gene regulatory function in vertebrates.

AN OUTBREAK OF FELINE INFECTIOUS PERITONITIS IN THREE RELATED SAND CATS (FELIS MARGARITA) IN HUMAN CARE.

Feline infectious peritonitis (FIP) is a systemic disease in felid species caused by infection with mutated forms of feline coronavirus (FCoV), and outbreaks can devastate exotic felid populations in human care. Feline infectious peritonitis was diagnosed in three of four related juvenile sand cats (Felis margarita) from a single institution over a 6-wk period. Case 1 was a 7-mon-old male found deceased with no premonitory signs. Case 2, an 8-mon-old male (littermate to Case 1), and Case 3, a 6-mon-old male (from a different litter with identical parentage), were evaluated for lethargy and anorexia 1 mon after Case 1. Both exhibited transient anisocoria and progressive lethargy, anorexia, and dehydration despite antibiotic and supportive treatment. Approximately 1 wk after initial presentation, Case 2 was humanely euthanized, and Case 3 was found deceased. Necropsy findings included intrathoracic and/or intra-abdominal lymphadenopathy (3/3 cases), bicavitary effusion (2/3), multifocal tan hepatic and intestinal nodules (1/3), and multifocal yellow renal nodules (1/3). Histologically, all cats had severe pyogranulomatous vasculitis in multiple organs, and the presence of FCoV antigen was confirmed using immunohistochemical staining. Next-generation sequencing of the virus from Case 3's affected kidney demonstrated ∼93% homology to the UG-FH8 virus, a serotype 1 feline alphacoronavirus isolated from Denmark. Future research will focus on comparative viral genomic sequencing with the goals of identifying potential sources of FCoV infection and identifying features that may have contributed to the development of FIP in this species.

Chemical crosslinking enhances RNA immunoprecipitation for efficient identification of binding sites of proteins that photo-crosslink poorly with RNA.

In eukaryotic cells, proteins that associate with RNA regulate its activity to control cellular function. To fully illuminate the basis of RNA function, it is essential to identify such RNA-associated proteins, their mode of action on RNA, and their preferred RNA targets and binding sites. By analyzing catalogs of human RNA-associated proteins defined by ultraviolet light (UV)-dependent and -independent approaches, we classify these proteins into two major groups: (i) the widely recognized RNA binding proteins (RBPs), which bind RNA directly and UV-crosslink efficiently to RNA, and (ii) a new group of RBP-associated factors (RAFs), which bind RNA indirectly via RBPs and UV-crosslink poorly to RNA. As the UV crosslinking and immunoprecipitation followed by sequencing (CLIP-seq) approach will be unsuitable to identify binding sites of RAFs, we show that formaldehyde crosslinking stabilizes RAFs within ribonucleoproteins to allow for their immunoprecipitation under stringent conditions. Using an RBP (CASC3) and an RAF (RNPS1) within the exon junction complex (EJC) as examples, we show that formaldehyde crosslinking combined with RNA immunoprecipitation in tandem followed by sequencing (xRIPiT-seq) far exceeds CLIP-seq to identify binding sites of RNPS1. xRIPiT-seq reveals that RNPS1 occupancy is increased on exons immediately upstream of strong recursively spliced exons, which depend on the EJC for their inclusion.

Inhibition of cytoplasmic cap methylation identifies 5' TOP mRNAs as recapping targets and reveals recapping sites downstream of native 5' ends.

Cap homeostasis is the cyclical process of decapping and recapping that maintains the translation and stability of a subset of the transcriptome. Previous work showed levels of some recapping targets decline following transient expression of an inactive form of RNMT (ΔN-RNMT), likely due to degradation of mRNAs with improperly methylated caps. The current study examined transcriptome-wide changes following inhibition of cytoplasmic cap methylation. This identified mRNAs with 5'-terminal oligopyrimidine (TOP) sequences as the largest single class of recapping targets. Cap end mapping of several TOP mRNAs identified recapping events at native 5' ends and downstream of the TOP sequence of EIF3K and EIF3D. This provides the first direct evidence for downstream recapping. Inhibition of cytoplasmic cap methylation was also associated with mRNA abundance increases for a number of transcription, splicing, and 3' processing factors. Previous work suggested a role for alternative polyadenylation in target selection, but this proved not to be the case. However, inhibition of cytoplasmic cap methylation resulted in a shift of upstream polyadenylation sites to annotated 3' ends. Together, these results solidify cap homeostasis as a fundamental process of gene expression control and show cytoplasmic recapping can impact regulatory elements present at the ends of mRNA molecules.

ROR1-targeted delivery of miR-29b induces cell cycle arrest and therapeutic benefit in vivo in a CLL mouse model.

Chronic lymphocytic leukemia (CLL) occurs in 2 major forms: aggressive and indolent. Low miR-29b expression in aggressive CLL is associated with poor prognosis. Indiscriminate miR-29b overexpression in the B-lineage of mice causes aberrance, thus warranting the need for selective introduction of miR-29b into B-CLL cells for therapeutic benefit. The oncofetal antigen receptor tyrosine kinase orphan receptor 1 (ROR1) is expressed on malignant B-CLL cells, but not normal B cells, encouraging us with ROR1-targeted delivery for therapeutic miRs. Here, we describe targeted delivery of miR-29b to ROR1+ CLL cells leading to downregulation of DNMT1 and DNMT3A, modulation of global DNA methylation, decreased SP1, and increased p21 expression in cell lines and primary CLL cells in vitro. Furthermore, using an Eµ-TCL1 mouse model expressing human ROR1, we report the therapeutic benefit of enhanced survival via cellular reprograming by downregulation of DNMT1 and DNMT3A in vivo. Gene expression profiling of engrafted murine leukemia identified reprogramming of cell cycle regulators with decreased SP1 and increased p21 expression after targeted miR-29b treatment. This finding was confirmed by protein modulation, leading to cell cycle arrest and survival benefit in vivo. Importantly, SP1 knockdown results in p21-dependent compensation of the miR-29b effect on cell cycle arrest. These studies form a basis for leukemic cell-targeted delivery of miR-29b as a promising therapeutic approach for CLL and other ROR1+ B-cell malignancies.

Single nucleotide polymorphisms affect RNA-protein interactions at a distance through modulation of RNA secondary structures.

Single nucleotide polymorphisms are widely associated with disease, but the ways in which they cause altered phenotypes are often unclear, especially when they appear in non-coding regions. One way in which non-coding polymorphisms could cause disease is by affecting crucial RNA-protein interactions. While it is clear that changing a protein binding motif will alter protein binding, it has been shown that single nucleotide polymorphisms can affect RNA secondary structure, and here we show that single nucleotide polymorphisms can affect RNA-protein interactions from outside binding motifs through altered RNA secondary structure. By using a modified version of the Vienna Package and PAR-CLIP data for HuR (ELAVL1) in humans we characterize the genome-wide effect of single nucleotide polymorphisms on HuR binding and show that they can have a many-fold effect on the affinity of HuR binding to RNA transcripts from tens of bases away. We also find some evidence that the effect of single nucleotide polymorphisms on protein binding might be under selection, with the non-reference alleles tending to make it harder for a protein to bind.

Maturation of 23S rRNA includes removal of helix H1 in many bacteria.

In most bacteria, the three ribosomal RNAs (rRNAs) are encoded together in each of several near-identical operons. As soon as the nascent precursor rRNA emerges from RNA polymerase, ribosome assembly begins. This process entails ribosomal protein binding, rRNA folding, rRNA modification, and rRNA processing. In the model organisms Escherichia coli and Bacillus subtilis, rRNA processing results in similar mature rRNAs, despite substantial differences in the cohort of RNAses involved. A recent study of Flavobacterium johnsoniae, a member of the phylum Bacteroidota (formerly Bacteroidetes), revealed that helix H1 of 23S rRNA is absent from ribosomes, apparently a consequence of rRNA maturation. In this work, we mined RNA-seq data from 19 individual organisms and ocean metatranscriptomic samples to compare rRNA processing across diverse bacterial lineages. We found that mature ribosomes from multiple clades lack H1, and typically these ribosomes also lack an encoded H98. For all groups analysed, H1 is predicted to form in precursor rRNA as part of a longer leader-trailer helix. Hence, we infer that evolutionary loss of H98 sets the stage for H1 removal during 50S subunit maturation.

DNA sequence influences hexasome orientation to regulate DNA accessibility.

Nucleosomes, the fundamental organizing units of eukaryotic genomes, contain ∼146 base pairs of DNA wrapped around a histone H3-H4 tetramer and two histone H2A-H2B dimers. Converting nucleosomes into hexasomes by removal of a H2A-H2B dimer is an important regulatory event, but its regulation and functional consequences are not well-understood. To investigate the influence of hexasomes on DNA accessibility, we used the property of the Widom-601 Nucleosome Positioning Sequence (NPS) to form homogeneously oriented hexasomes in vitro. We find that DNA accessibility to transcription factors (TF) on the hexasome H2A-H2B distal side is identical to naked DNA, while the accessibility on the H2A-H2B proximal side is reduced by 2-fold, which is due to a 2-fold reduction in hexasome unwrapping probability. We then determined that a 23 bp region of the Widom-601 NPS is responsible for forming homogeneously oriented hexasomes. Analysis of published ChIP-exo data of hexasome containing genes identified two DNA sequence motifs that correlate with hexasome orientation in vivo, while ExoIII mapping studies of these sequences revealed they generate homogeneously oriented hexasomes in vitro. These results indicate that hexasome orientation, which is influenced by the underlying DNA sequence in vivo, is important for modulating DNA accessibility to regulate transcription.

Global analysis of protein synthesis in Flavobacterium johnsoniae reveals the use of Kozak-like sequences in diverse bacteria.

In all cells, initiation of translation is tuned by intrinsic features of the mRNA. Here, we analyze translation in Flavobacterium johnsoniae, a representative of the Bacteroidetes. Members of this phylum naturally lack Shine-Dalgarno (SD) sequences in their mRNA, and yet their ribosomes retain the conserved anti-SD sequence. Translation initiation is tuned by mRNA secondary structure and by the identities of several key nucleotides upstream of the start codon. Positive determinants include adenine at position -3, reminiscent of the Kozak sequence of Eukarya. Comparative analysis of Escherichia coli reveals use of the same Kozak-like sequence to enhance initiation, suggesting an ancient and widespread mechanism. Elimination of contacts between A-3 and the conserved ß-hairpin of ribosomal protein uS7 fails to diminish the contribution of A-3 to initiation, suggesting an indirect mode of recognition. Also, we find that, in the Bacteroidetes, the trinucleotide AUG is underrepresented in the vicinity of the start codon, which presumably helps compensate for the absence of SD sequences in these organisms.

The role of annealing and fragmentation in human tau aggregation dynamics.

Alzheimer's disease pathogenesis is associated with the conversion of monomeric tau protein into filamentous aggregates. Because both toxicity and prion-like spread of pathogenic tau depend in part on aggregate size, the processes that underlie filament formation and size distribution are of special importance. Here, using a combination of biophysical and computational approaches, we investigated the fibrillation dynamics of the human tau isoform 2N4R. We found that tau filaments engage in a previously uncharacterized secondary process involving end-to-end annealing and that rationalization of empirical aggregation data composed of total protomer concentrations and fibril length distributions requires inclusion of this process along with filament fragmentation. We noted that annealing of 2N4R tau filaments is robust, with an intrinsic association rate constant of a magnitude similar to that mediating monomer addition and consistent with diffusion-mediated protein-protein interactions in the absence of long-range attractive forces. In contrast, secondary nucleation on the surface of tau filaments did not detectably contribute to tau aggregation dynamics. These results indicate that tau filament ends engage in a range of homotypic interactions involving monomers, oligomers, and filaments. They further indicate that, in the case of tau protein, fibril annealing and fragmentation along with primary nucleation and elongation are the major processes controlling filament size distribution.