Determining jumping performance from a single body-worn accelerometer using machine learning.

External peak power in the countermovement jump is frequently used to monitor athlete training. The gold standard method uses force platforms, but they are unsuitable for field-based testing. However, alternatives based on jump flight time or Newtonian methods applied to inertial sensor data have not been sufficiently accurate for athlete monitoring. Instead, we developed a machine learning model based on characteristic features (functional principal components) extracted from a single body-worn accelerometer. Data were collected from 69 male and female athletes at recreational, club or national levels, who performed 696 jumps in total. We considered vertical countermovement jumps (with and without arm swing), sensor anatomical locations, machine learning models and whether to use resultant or triaxial signals. Using a novel surrogate model optimisation procedure, we obtained the lowest errors with a support vector machine when using the resultant signal from a lower back sensor in jumps without arm swing. This model had a peak power RMSE of 2.3 W·kg-1 (5.1% of the mean), estimated using nested cross validation and supported by an independent holdout test (2.0 W·kg-1). This error is lower than in previous studies, although it is not yet sufficiently accurate for a field-based method. Our results demonstrate that functional data representations work well in machine learning by reducing model complexity in applications where signals are aligned in time. Our optimisation procedure also was shown to be robust can be used in wider applications with low-cost, noisy objective functions.

Identification of linear surface epitopes on the guinea pig sperm membrane protein PH-20.

The sperm plasma membrane protein PH-20 has previously been shown to be an effective immunogen for protection against fertilization in guinea pigs. To identify immunodominant regions on gpPH-20 that may be related to this contraceptive effect, we used several high-titer immune sera obtained from animals rendered infertile by gpPH-20 injections to screen a set of overlapping peptides that cover the entire 494-residue sequence. Multiple clusters of peptide sequences exhibited specific reactivity. Some of these sequences were then constructed as octameric synthetic peptides and tested for immunogenicity in female guinea pigs. Our results indicated two regions (res. 94-119 and res. 424-444) to be highly immunogenic and both are surface accessible when native gpPH-20 is in solution or anchored on sperm surface. Both anti-peptide antibodies are specific for gpPH-20 and one of them inhibited hyaluronidase activity partially. These monospecific antibodies should be useful probes for further molecular definition of gpPH-20 structure-function relationships.

Recombinant human monoclonal antibody IgG1b12 neutralizes diverse human immunodeficiency virus type 1 primary isolates.

The CD4-binding domain of human immunodeficiency virus type 1 (HIV-1) gp120 elicits antibodies that are present in infected human sera. Monoclonal antibodies that recognize the HIV-1 gp120 CD4-binding domain have been isolated. Some of these antibodies can neutralize laboratory-adapted strains of HIV-1 and probably mediate neutralization by interfering with virus binding to its cellular CD4 receptor. However, most anti-CD4 binding domain antibodies do not neutralize primary HIV-1 isolates. We used primary HIV-1 isolates in an infectivity reduction assay to test the uniquely derived anti-CD4 binding domain recombinant human monoclonal antibody, IgG1b12. All of the tested HIV-1 isolates were neutralized by this antibody. Additional studies indicated that neutralization of a primary isolate with MAb IgG1b12 did not require continuous exposure of human peripheral blood mononuclear cell cultures to the antibody. Finally, a complete IgG1 molecule of an in vitro-selected b12 FAb mutant with a > 400-fold increase in affinity was assembled, expressed in mammalian cells, and evaluated in the infectivity reduction assay in comparative studies with the parent IgG1b12 antibody. The mutant did not retain the level of primary isolate neutralization potency that was a property of the parent molecule. Thus, we confirm that recombinant IgG1b12 has a unique specificity, and that it can neutralize all primary isolates tested in human PBMC cultures in vitro.

A neutralizing epitope of human papillomavirus type 11 is principally described by a continuous set of residues which overlap a distinct linear, surface-exposed epitope.

A panel of monoclonal antibodies (MAbs) which neutralize human papillomavirus type 11 (HPV11) in the athymic mouse xenograph neutralization assay and bind HPV11 virus-like particles (VLPs) has been described. We recently presented evidence that the Gly131-Tyr132 residues of the major capsid protein L1 confer type 11-specific binding. However, residues distally located on the primary L1 sequence also were shown to affect binding. This poses the question whether the epitope is principally centered in the region of Gly131-Tyr132 or, alternatively, is comprised of diversely located residues which come into proximity only upon proper assembly. We analyzed the result of numerous substitutions located between Tyr123 and Val142 of the HPV11 L1 sequence. We show that substitutions at five positions result in loss of binding for one or more of these MAbs by an enzyme-linked immunosorbent assay which measures antibody binding to VLPs. We demonstrate that binding of these MAbs is redirected to HPV16 VLPs which harbor eight type 11-like substitutions within the homologous region. Three of these substitutions did not affect binding when individually substituted in HPV11 but yet were still required to transfer binding to substituted HPV16 VLPs. The results demonstrate that the epitope for this class of neutralizing MAbs, although conformational and requiring VLP assembly for presentation, principally lies along a 20-residue stretch of the L1 major capsid protein. This targets the region for evaluation of the possibility of receptor binding and suggests possibilities for the design of peptide inhibitors of virus infectivity.

The consequence of passive administration of an anti-human immunodeficiency virus type 1 neutralizing monoclonal antibody before challenge of chimpanzees with a primary virus isolate.

The anti-gp41 virus neutralizing monoclonal antibody 2F5 was infused into chimpanzees, which were then given an intravenous challenge with a primary human immunodeficiency virus type I (HIV-1) isolate. In two control animals, the infection was established immediately, as evidenced by positive cell-associated DNA PCR and serum RNA PCR tests within 1 week, seroconversion by 4 weeks, and development of lymphadenopathy in this acute phase. Serum RNA PCR tests were negative in one of the two antibody-infused animals until week 8 and in the other antibody-infused animal until week 12; both animals seroconverted at week 14. The peak of measurable virus-specific serum RNA was delayed until week 16 in one antibody-infused animal. Virus-specific RNA in the other animal did not reach levels comparable to those in the other animals through 1 year of follow-up studies. Virus was isolated from the week 16 blood sample from one infused animal. Virus was not isolated from peripheral blood of the second animal but was isolated from lymph node cells taken at week 36. The infection of untreated chimpanzees with this primary isolate appears robust. Use of this isolate should widen the scope of possible experiments in the chimpanzee model. This antibody infusion study indicates that neutralizing antibody, when present at the time of challenge, affects the timing and level of infection and remains influential after it can no longer be detected in the peripheral circulation. It is possible that preexisting, neutralizing antibodies (passively administered or actively elicited) affect the course of acute-phase virus replication in humans. It remains to be established whether these immunologically mediated early effects will influence the course of HIV-1 disease.

Molecular cloning of the human fertilin beta subunit.

Fertilin, a heterodimeric sperm surface protein, first identified in guinea pig, has been shown to play an important role in sperm-egg interactions. We report here the complementary DNA and deduced amino acid sequence of human fertilin beta. The human fertilin beta shares significant sequence homology with mouse, guinea pig and monkey fertilin beta and also exhibits similar structural organization. Of particular interest, the mature guinea pig fertilin beta contains an amino-terminal 90 amino acid disintegrin domain. It has been suggested that the integrin recognition sequence (TDE) of the guinea pig fertilin beta disintegrin domain mediates sperm-egg binding. The amino acid sequence at this position in human fertilin beta differs from the mouse, guinea pig and monkey sequence (mouse-QDE; human-FEE; monkey-FDE).

Two amino acid residues confer type specificity to a neutralizing, conformationally dependent epitope on human papillomavirus type 11.

Characterization of virus binding by neutralizing antibodies is important both in understanding early events in viral infectivity and in development of vaccines. Neutralizing monoclonal antibodies (MAbs) to human papillomavirus type 11 (HPV11) have been described, but mapping the binding site has been difficult because of the conformational nature of key type-specific neutralization epitopes on the L1 coat protein. We have determined those residues of the L1 protein of HPV11 which confer type specificity to the binding of HPV11-neutralizing MAbs. Binding of three HPV11-specific neutralizing MAbs could be redirected to HPV6 L1 virus-like particles in which as few as two substitutions of corresponding amino acid residues from HPV11 L1 have been made, thus demonstrating the importance of these residues to MAb binding through the transfer of a conformationally dependent epitope. In addition, a fourth neutralizing MAb could be distinguished from the other neutralizing MAbs in terms of the amino acid residues which affect binding, suggesting the possibility that it neutralizes HPV11 through a different mechanism.

Rapid, high-level transient expression of papillomavirus-like particles in insect cells.

Empirical scanning of natural or engineered peptide sequences for functional residues is inherently dependent upon efficient expression of large numbers of individual sequence variants to assay their relative functional potency. The insect baculovirus system has been widely used for expression of viral coat proteins, but it generally requires prior isolation and expansion of a plaque-purified recombinant viral stock to generate useful quantities of self-assembled virus-like particles. In search of a more rapid means of expression of analytical levels of the L1 coat protein of cottontail rabbit and human type 11 papilloma-viruses, we found that even brief transient cotransfection of insect cells with baculovirus plasmid transfer vectors and viral DNA yielded assembled particles that were immunologically indistinguishable from particles obtained with plaque-purified viral stocks. Within six days of plasmid/viral DNA cotransfection of Sf9 cells, at least 1-2 micrograms of assembled L1 particles/100-mm plate could be demonstrated, which proved more than sufficient to assay functionality. Transient cotransfection of insect cells should provide general utility for rapid high-level expression of sets of protein sequence variants, as well as other sequence-scanning applications such as sequence optimization in protein engineering.

Use of a novel mutagenesis strategy, optimized residue substitution, to decrease the off-rate of an anti-gp120 antibody.

We have developed a novel strategy to decrease the antibody:antigen off-rate which we call optimized residue substitution. This strategy employs alanine substitution to first identify residues non-optimal for binding, as evidenced by a decrease in off-rate upon alanine replacement. These positions are then individually randomized to all amino acids, and the best replacement for each position determined. Finally, a construct which combines all optimized substitutions is generated and evaluated. We applied this strategy to the heavy chain CDR3 of P5Q, a scFv antibody which recognizes an epitope on the V3 loop of HIV gp120. We identified two amino acid substitutions that together decrease the off-rate by nearly ten-fold. The contributions by the two substitutions were near additive, indicative of independent affects on binding. We suggest that this strategy can be generalized to strengthen protein:ligand and protein:protein interactions in other systems.

Enhancing the avidity of a human recombinant anti-HIV-1 monoclonal antibody through oligomerization.

The oligomerization by chemical cross-linking of a recombinant human antiviral monoclonal antibody (MAb), r447-1, and its characterization are described. This MAb binds to an epitope residing in the hypervariable V3 region of the envelope protein (gp120/160) of HIV-1. A dimeric form of this MAb displays enhanced avidity and was found to be capable of neutralizing a greater variety of lymphoid cell culture-adapted HIV-1 variants and HIV-1 primary isolates than its monomeric form. The superior binding and breadth of reactivity of this antibody suggests it may have utility as a therapeutic and/or prophylactic agent, if it possesses an appropriate safety and immunogenicity profile.

Sequence of the dog immunoglobulin alpha and epsilon constant region genes.

The immunoglobulin alpha (IGHAC) and epsilon (IGHEC) germline constant region genes were isolated from a dog liver genomic DNA library. Sequence analysis indicates that the dog IGHEC gene is encoded by four exons spread out over 1.7 kilobases (kb). The IGHAC sequence encompasses 1.5 kb and includes all three constant region coding exons. The complete exon/intron sequence of these genes is described.

A microplate assay for hyaluronidase and hyaluronidase inhibitors.

A sensitive, high-throughput assay has been developed which measures hyaluronidase activity in a microplate format. In this assay, hyaluronic acid is suspended in agarose in a microtiter well, the plate is incubated with the hyaluronidase solution, and the undigested hyaluronic acid is precipitated with cetylpyridinium chloride. The precipitate blocks light transmittance and therefore an increase in the visible light transmitted correlates with the amount of digested hyaluronic acid. Using this assay, as little as 0.05 units of hyaluronidase activity can be detected. The assay is highly reproducible and can be run with commercially available reagents. Hyaluronidase activity is easy to quantitate using a microplate reader and this format allows large numbers of samples to be assayed.

Structural characterization of the N-glycans of a humanized anti-CD18 murine immunoglobulin G.

This study characterized the N-glycans of a humanized immunoglobulin G4 (IgG4) expressed in NS/O mouse myeloma cells and directed against the CD18 family of adhesion-promoting receptors on leukocytes. The N-glycans were released from the purified recombinant IgG by N-glycanase treatment, purified by Sephadex G50 chromatography, and fractionated by Bio-Gel P-4 chromatography into three oligosaccharide pools. Each pool was analyzed individually by glycosyl composition analysis, high-pH anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD), 600-MHz 1H-NMR spectroscopy, and electrospray-ionization mass spectrometry. In addition, each of the three pools was subfractionated by HPAEC and the isolated subfractions that contained sufficient material were hydrolyzed and analyzed for glycosyl composition by HPAEC-PAD. The overall results indicate the presence of five oligomannoside-type structures (containing 5 to 8 Man residues) which are not usually found in IgG, and the presence of eight diantennary (mostly truncated) N-acetyllactosamine-type structures which are typical of mouse and human IgGs. The N-acetyllactosamine-type structures were heterogeneous with regard to alpha(1-->6) fucosylation of the linkage GlcNAc, and the presence or absence of GlcNAc and/or Gal beta(1-->4)GlcNAc extending the core pentasaccharide (Man3GlcNAc2). No evidence was found for the presence of sialic acid or bisecting GlcNAc residues on the N-acetyllactosamine-type chains. The latter finding suggests that the N-glycans of this humanized IgG are of the mouse type.

Activation of the c-Raf protein kinase by protein kinase C phosphorylation.

The product of the c-raf-1 proto-oncogene is a cytoplasmic serine/threonine protein kinase that appears to be activated in signal transduction from a variety of cell-surface receptors. The mechanism of c-Raf activation upon stimulation of cell-surface receptors is not clear, but there seem to exist multiple pathways of activation which involve tyrosine and/or serine phosphorylation of the c-Raf protein in vivo. The activated state of Raf is reflected in an increased apparent molecular weight of the Raf protein in sodium dodecyl sulfate-polyacrylamide gels owing to hyperphosphorylation. The tumor promoter 12-O-tetradecanoyl phorbol 13-acetate (TPA) is one of the agents able to induce this hyperphosphorylation of Raf in vivo, suggesting that protein kinase C (PKC) may be involved in the activation of c-Raf in particular situations. Using recombinant baculoviruses expressing PKC and Raf polypeptides, we show here that conventional PKC types (alpha, beta, gamma) but not novel types (delta, zeta, eta) or the unrelated Mos kinase are able to activate c-Raf in a TPA-dependent manner upon coexpression in insect cells. Direct phosphorylation of the Raf protein with PKC in vitro also enhanced the kinase activity of c-Raf, suggesting that c-Raf acts immediately downstream of PKC in a protein kinase cascade which is triggered by TPA and may lead to transcriptional activation of TPA-inducible genes and tumor promotion.

Expression of recombinant human anti-MAG antibodies in non-lymphoid mammalian cells.

The variable heavy and light chain genes of a monoclonal, IgM, anti-MAG antibody from a patient with neuropathy were inserted into expression vectors containing the gamma and kappa constant regions respectively and co-transfected into monkey kidney CV1P cells. The expressed antibody had the same antigenic specificity but significantly lower avidity than the native IgM, anti-MAG, antibody as detected by ELISA. When the variable heavy chain gene of the anti-MAG antibody was co-transfected with the variable light chain gene from another monoclonal, IgM, anti-MAG antibody, a fully assembled antibody was expressed as determined by a trapping ELISA, but it did not bind to (MAG) or to sulfated glucuronic acid paragloboside, indicating that both heavy and light chains contribute to the binding activity.

Endogenous loading of HLA-A2 molecules with an analog of the influenza virus matrix protein-derived peptide and its inhibition by an exogenous peptide antagonist.

Episomal plasmids (p8901) with minigenes coding for the influenza virus matrix peptide amino acids 57-68 (KGILGFVFTLTV; referred to as M57-68) or coding for a modified peptide were introduced into HLA-A2-positive target cells. The association of these peptides, synthesized in the cytoplasm, with HLA-A2 and the expression of this complex at the cell surface was evaluated with HLA-A2-restricted CTL specific for the influenza virus matrix peptide M57-68. Cells expressing M57-68 were lysed effectively, as were cells expressing a peptide that retained residues 60-64 with seven flanking alanine residues (AAALGFVFAAAA). An exogenously added synthetic analog of peptide M57-68 that inhibited sensitization of targets with synthetic peptide M57-68 also inhibited lysis of cells expressing the minigene coding for the peptide with seven alanine substitutions. These results demonstrate the utility of minigene DNA constructs in creating experimental systems to develop agents to diminish the severity of CTL-mediated tissue damage in autoimmune diseases and graft rejection.

A single amino acid substitution in a common African allele of the CD4 molecule ablates binding of the monoclonal antibody, OKT4.

The CD4 molecule is a relatively non-polymorphic 55 kDa glycoprotein expressed on a subset of T lymphocytes. A common African allele of CD4 has been identified by non-reactivity with the monoclonal antibody, OKT4. The genetic basis for the OKT4- polymorphism of CD4 is unknown. In the present paper, the structure of the CD4 molecule from an homozygous CD4OKT4- individual was characterized at the molecular level. The size of the CD4OKT4- protein and mRNA were indistinguishable from those of the OKT4+ allele. The polymerase chain reaction (PCR) was used to map the structure of CD4OKT4- cDNAs by amplifying overlapping DNA segments and to obtain partial nucleotide sequence after asymmetric amplification. PCR was then used to clone CD4OKT4- cDNAs spanning the coding region of the entire, mature CD4 protein by amplification of two overlapping segments followed by PCR recombination. The nucleotide sequence of CD4OKT4- cDNA clones revealed a G----A transition at bp 867 encoding an arginine----tryptophan substitution at amino acid 240 relative to CD4OKT4+. Expression of a CD4OKT4- cDNA containing only this transition, confirmed that the arginine----tryptophan substitution at amino acid 240 ablates the binding of the mAb OKT4. A positively charged amino acid residue at this position is found in chimpanzee, rhesus macaque, mouse and rat CD4 suggesting that this mutation may confer unique functional properties to the CD4OKT4- protein.

Involvement of the RAF1 locus, at band 3p25, in the 3p deletion of small-cell lung cancer.

The ability to establish long-term cell lines of small-cell lung cancer (SCLC) has provided an in vitro model for the disease. We report on the characterization of 10 new human SCLC cell lines established from 34 cytopathologically positive specimens. Based on morphologic and biochemical characterization, growth properties, and expression of MYC and neuroendocrine properties, eight cell lines were categorized as "classic" and two cell lines as "variant". Cytogenetic examination revealed loss of all or part of 3p in all nine SCLC cell lines analyzed. The smallest deletion in common was found at 3p21-3p25. Restriction fragment length polymorphism (RFLP) analyses with probes for 3p were performed for correlation with karyotypic data and supported the cytogenetic findings. In 21 SCLC specimens (cell lines and tumor tissue) with normal DNA, used for comparison, we observed loss of heterozygosity at RAF1 (3p25) in ten of ten informative pairs by using two RFLPs from the RAF1 locus. In addition, loss of heterozygosity was noted in nine of 10 pairs examined with DNF15S2 (3p21) and four of four with D3S3 (3p14). Analysis of cell lines and tumor specimens that lacked paired normal tissue showed a homozygous pattern with the RAF1 probes in all 18 cases. Northern blots revealed significant expression of RAF1 in all cell lines tested. The transcript size was normal. The cytogenetic and RFLP data suggest that the RAF1 locus at 3p25 is involved in the chromosomal deletion of SCLC.

Effective thrombolysis without marked plasminemia after bolus intravenous administration of vampire bat salivary plasminogen activator in rabbits.

BACKGROUND: The use of recombinant tissue-type plasminogen activator (t-PA) in thrombolytic therapy is frequently associated with significant fibrinogenolysis. In contrast, recombinant vampire bat salivary plasminogen activator (Bat-PA) displays strict fibrin specificity, an attribute that could be desirable in a fibrinolytic agent. METHODS AND RESULTS: The efficacy and fibrin selectivity of Bat-PA was evaluated and compared with that of t-PA using a rabbit model of femoral arterial thrombosis. Administration of 8.1, 14, and 42 nmol Bat-PA/kg by bolus intravenous injection restored flow in 50%, 75%, and 80% of the rabbits, respectively. The incidence of reperfusion after bolus intravenous injection of 14 and 42 nmol t-PA/kg was 15% and 78%, respectively. The maximal femoral artery reperfusion flows were equivalent after treatment with 42 nmol Bat-PA/kg or 42 nmol t-PA/kg, but the time to reach maximal flow for Bat-PA was approximately one half that of t-PA. Furthermore, the rapid restoration of flow by 42 nmol Bat-PA/kg, in contrast to equimolar t-PA, was accomplished without fibrinogenolysis and with only small decreases in the plasminogen and alpha 2-antiplasmin levels. Equipotent doses of Bat-PA and t-PA both resulted in approximate 2.5-fold increases in the template bleeding times of aspirin-pretreated rabbits. The clearance of Bat-PA from rabbits exhibited biexponential elimination kinetics; approximately 80% was cleared by the relatively slow beta phase (half-life of 17.1 minutes). Overall, Bat-PA was cleared approximately fourfold slower than t-PA. CONCLUSIONS: Bolus intravenous administration of Bat-PA would facilitate prompt initiation of thrombolytic therapy, and the avoidance of plasminemia could result in fewer and less severe bleeding complications.