Role of βγ-Dimers of GTP-Binding Proteins in Processes of Hormonal Signal Transduction

This review summarizes and analyzes data on structural and functional peculiarities of - and -subunits of heterotrimeric G proteins and their participation in -dimer in processes of hormonal signal transduction to the cell. The molecular mechanisms are discussed, which are responsible for formation of the -dimer complex and its interaction with the wide spectrum of signal and structural proteins, such as G protein -subunit, serpentine type receptors, kinases of G protein-coupled receptors, adenylyl cyclase, phospholipase C2, phosphatidyl-inositol-3-kinase, Raf-kinase, potassium and calcium channels, microtubular and RGS-proteins. The characteristics of - and -subunits are given for living organisms of different phylogenetic levels (lower eukaryotes, fungi, plants, vertebrates, and invertebrates). It is concluded that the -dimer is the most important regulatory link in various signal systems to carry out switch from one signal cascade to another and thereby to determine the final cellular response to the signal impulse.     

Aldimine to ketoamine isomerization (Amadori rearrangement) potential at the individual nonenzymic glycation sites of hemoglobin a: Preferential inhibition of glycation by nucleophiles at sites of low isomerization potential

The relative roles of the two structural aspects of nonenzymic glycation sites of hemoglobin A, namely the ease with which the amino groups could form the aldimine adducts and the propensity of the microenvironments of the respective aldimines to facilitate the Amadori rearrangement, in dictating the site selectivity of nonenzymic glycation with aldotriose has been investigated. The chemical reactivity of the amino groups of hemoglobin A forin vitro reductive glycation with aldotriose is distinct from that in the nonreductive mode. The reactivity of amino groups of hemoglobin A toward reductive glycation (i.e., propensity for aldimine formation) decreases in the order Val-1(), Val-1(), Lys-66(), Lys-61(), and Lys-16(). The overall reactivity of hemoglobin A toward nonreductive glycation decreased in the order Lys-16(), Val-1(), Lys-66(), Lys-82(), Lys-61(), and Val-1(). Since the aldimine is the common intermediate for both the reductive and nonreductive modification, the differential selectivity of protein for the two modes of glycation is clearly a reflection of the propensity of the microenvironments of nonenzymic glycation sites to facilitate the isomerization reaction (i.e., Amadori rearrangement). A semiquantitative estimate of this propensity of the microenvironment of the nonenzymic glycation sites has been obtained by comparing the nonreductive (nonenzymic) and reductive modification at individual glycation sites. The microenvironment of Lys-16() is very efficient in facilitating the rearrangement and the relative efficiency decreases in the order Lys-16(), Lys-82(), Lys-66(), Lys-61(), Val-1(), and Val-1(). The propensity of the microenvironment of Lys-16() to facilitate the Amadori rearrangement of the aldimine is about three orders of magnitude higher than that of Val-1() and is about 50 times higher than that of Val-1(). The extent of nonenzymic glycation at the individual sites is modulated by various factors, such as thepH, concentration of aldotriose, and the concentration of the protein. The nucleophiles—such as tris, glycine ethyl ester, and amino guanidine—inhibit the glycation by trapping the aldotriose. The nonenzymic glycation inhibitory power of nucleophile is directly related to its propensity to form aldimine. Thus, the extent of inhibition of nonenzymic glycation at a given site by a nucleophile directly reflects the relative role ofpK _a of the site in dictating the glycation at that site. The nonenzymic glycation of an amino group of a protein is an additive/synergestic consequence of the propensity of the site to form aldimine adducts on one hand, and the propensity of its microenvironment to facilitate the isomerization of the aldimines to ketoamines on the other. The isomerization potential of microenvironment plays the dominant role in dictating the site specificity of the nonenzymic glycation of proteins.     

13Cα and 13Cβ chemical shifts as a tool to delineate β-hairpin structures in peptides

Unravelling the factors that contribute to the formation and the stability of -sheet structure in peptides is a subject of great current interest. A -hairpin, the smallest -sheet motif, consists of two antiparallel hydrogen-bonded -strands linked by a loop region. We have performed a statistical analysis on protein -hairpins showing that the most abundant types of -hairpins, 2:2, 3:5 and 4:4, have characteristic patterns of ¹³C and ¹³C conformational shifts, as expected on the basis of their and angles. This fact strongly supports the potential value of ¹³C and ¹³C conformational shifts as a means to identify -hairpin motifs in peptides. Their usefulness was confirmed by analysing the patterns of ¹³C and ¹³C conformational shifts in 13 short peptides, 10–15 residues long, that adopt -hairpin structures in aqueous solution. Furthermore, we have investigated their potential as a method to quantify -hairpin populations in peptides.     

Zur Frage des Zeipunktes einer Pr?gung auf die Heimatregion beim Halsbandschn?pper(Ficedula albicollis)

Zusammenfassung Junge Halsbandschnäpper wurden handaufgezogen, flogen im Flugkäfig aus und wurden dort selbständig. Darauf wurden sie 90 km nach Süden verfrachtet und in einem von dieser Art unbewohnten Gebiet freigelassen. Im nächsten Frühjahr siedelten sich mindestens 9 dort an, was 19% Rückkehrern entspricht, wenn die Hälfte der Vögel waren. kehrten in geringerer Zahl zurück und wurden nicht restlos erfaßt.Eine weitere Gruppe wurde erst vor Ende der Jugendmauser verfrachtet. Auch davon kehrten 18-19% der zurück. Ein Zeitraum von rund 2 Wochen vor dem Wegzug reichte also zur Prägung auf ein Gebiet als Heimat aus.Von einer dritten Gruppe von insgesamt 68 Schnäppern (= ca. 34 ), die erst nach Ende der Jugendmauser zur Wegzugzeit aufgelassen wurde, konnte später keiner nachgewiesen werden, auch nicht am Aufzuchtsort. Letzteres könnte an der Ungunst der örtlichen Verhältnisse liegen.Mit Unterstützung der Deutschen Forschungsgemeinschaft.     

Enzyme-catalyzed synthesis of alkyl β-D-glycosides with crude homogenate ofSulfolobus solfataricus

Summary Crude homogenate of thermophilic archaebacteriumSulfolobus solfataricus, possessing a -glycosidase, has been used to synthesize different alkyl -D-glycosides starting from phenyl -D-glucoside, phenyl -D-galactoside and lactose as carbohydrate donors. High product yield (95% with respect to the carbohydrate donor) of octyl -D-glucoside has been obtained in a two-phase system containing 5% of water. The enantioselection for the galactosyl transfer to the secondary hydroxyl group of propane-1,2-diol is higher than that found using -galactosidase fromE. coli.     

Immunosurgical treatment of stage I malignant melanoma

Summary We treated thirty seven patients suffering from a clinical stage I malignant melanoma by immunosurgery and observed them for a three-year period ending December 1975. Prior to surgery the tumor was submitted to the application of DNCB. Patients were divided between those who had or had not been subjected to biopsy before starting treatment. Accordingly, 23 patients treated by immunosurgery as the first treatment, were called the primitive treatment group and 14 patients subjected to a previous biopsy, the secondary treatment group. In the primitive treatment group, all 23 patients survived without any sign of recurrence. Relapse was observed in 4 out of 14 patients of the biopsy group. Extensive histological investigations have been made by light and electron microscopes. The tumors were infiltrated and disrupted by a large number of mononucleated cells. The majority of infiltrating cells were small lymphocytes. True lymphatic follicles have been observed in the skin around the melanoma.     

A model of associative memory in the brain

A model of associative memory for time varying spatial patterns is proposed and simulated on a digital computer. This is a network composed of many neuron-like elements, and shows an ability for associative memory similar to that of the brain.Suppose a number of sequences of spatial patterns are presented to this network, for example, 12345, ABC, and so on. Then, these patterns are memorized in the network. After that, if any part of one of these sequences, say 23, is presented to the circuit, the rest of the sequence, 45, is recalled following to it. It resembles to such a situation — if we hear a part of a melody which we have memorized in the past, the rest of the melody is recalled even after it is stopped half-way. Although the recalled patterns are not always 100% correct, they are not completely destroyed even if the presented patterns are imperfect.     

Production of a novel syrup containing neofructo-oligosaccharides by the cells of Penicillium citrinum

A novel syrup containing neofructo-oligosaccharides was produced from sucrose (Brix 70) by whole cells of Penicillium citrinum. The efficiency of fructo-oligosaccharides production was more than 55% and those of the main carbohydrate components, 1-kestose (Fruf 21Fruf 21 Glc), nystose (Fruf 21Fruf 21 Fruf 21 Glc) and neokestose (Fruf 26 Glc12 Fruf), were 22, 14 and 11%, respectively.     

Effect of phosphate and adenine nucleotides on the rate of labeling of functional groups at the catalytic site of F1-ATPase

The effect of inorganic phosphate, ADP, ATP, and their analogues on the rate of labeling of F₁-ATPase by 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-Cl) and phenylglyoxal have been investigated. Analysis of the kinetic data indicate that the labeled functional groups of the essential tyrosine and arginine residues respectively are both located at the catalytic site of F₁. The active phenolic group of tyrosine is located closer to the bound inorganic phosphate or the -phosphate group than the - and -phosphate groups of the bound ATP at the catalytic site, whereas the guanidinium group of arginine is located closer to the - and -phosphate groups of the bound ATP than to its -phosphate group or the bound inorganic phosphate. The kinetically deduced dissociation constants are 1.3 mM and 210 µM for the inorganic phosphate and ADP respectively bound to this catalytic site. Labeling the essential tyrosine residue by NDB-Cl has been found to facilitate subsequent labeling of the essential arginine residue by phenylglyoxal.Abbreviations NBD-Cl 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole (this compound has been named 4-chloro-7-nitro-benzofurazan and abbreviated NBf-Cl elsewhere) - DTT dithiothreitol - EDTA ethylenediaminetetraacetic acid - P_i inorganic phosphate - PEP phosphoenolpyruvate - ADPCP ,-methylene-adenosine 5-triphosphate - AMPCP ,-methylene-adenosine 5-diphosphate - Hepes N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid - Tris 2-amino-2(hydroxymethyl)-1,3-propanediol     

Binding of Amyloid Beta-Protein to Intracellular Brain Proteins in Rat and Human

Amyloid beta-protein (A), in its soluble form, is known to bind several circulatory proteins such as apolipoprotein (apo) E, apo J and transthyretin. However, the binding of A to intracellular proteins has not been studied. We have developed an overlay assay to study A binding to intracellular brain proteins. The supernatants from both rat and human brains were found to contain several proteins that bind to A 1–40 and A 1–42. No major difference was observed in the A binding-proteins from brain supernatants of patients with Alzheimer's disease and normal age-matched controls. Binding studies using shorter amyloid beta-peptides and competitive overlay assays showed that the binding site of A to brain proteins resides between 12–28 amino acid sequence of A. The presence of several intracellular A-binding (AB) proteins suggests that these proteins may either protect A from its fibrillization or alternatively promote A polymerization. Identification of these proteins and their binding affinities for A are needed to assess their potential role in the pathogenesis of Alzheimer's disease.     

Ultrastructure of four types of striated muscle fibers in the atlantic hagfish (Myxine glutinosa,L.)

Summary Four types of striated muscle fibers with distinctive ultrastructure were defined in the Atlantic hagfish (Myxine glutinosa, L.): white, intermediate, and red fibers of m. parietalis, and red fibers of m. craniovelaris.White fibers are thick, contain very few mitochondria and fat vacuoles, and possess distinct and separate myofibrils with thin Z-disks and distinct M-lines. Intermediate fibers are thinner, possess largely similar myofibrils that often are even better separated due to a higher content of fat vacuoles and especially mitochondria and glycogen granules. Red fibers of m. parietalis contain large amounts of mitochondria, fat vacuoles, and glycogen granules. Their myofibrils possess M-lines, and although branching more, the myofibrils of red fibers conform with a Fibrillenstruktur pattern like those of white and intermediate fibers. Red fibers of m. craniovelaris are very thin, possess many smaller fat vacuoles, and large amounts of mitochondria and glycogen granules. The myofibrils are significantly thinner than in m. parietalis fibers, run as quite independent well separated units, possess thicker Z-disks, and lack M-lines. Large amounts of myosatellite cells are associated with these red fibers.Triads are located near A/I-junctions in all four fiber types and occur irregularly, the density of triads being different in the various fiber types.We are indebted to Dr. Finn Walvig, Biological Station, University of Oslo, Drøbak, for supply of hagfishes, and we also wish to thank Dr. Jan K. S. Jansen, Institute of Physiology, University of Oslo, for valuable suggestions during this study.     

UDP-GlcNAc: Galβ3GalNAc-Mucin: (GlcNAc → GalNAc) β6-N-Acetylglucosaminyltransferase and UDP-GlcNAc: Galβ3(GlcNAcβ6) GalNAc-Mucin (GlcNAc → Gal)β3-N-Acetylglucosaminyltransferase from Swine Trachea Epithelium

Summary Two specific -N-acetylglucosaminyltransferases involved in the branching and elongation of mucin oligosaccharide chains, namely, a 1,6 N-acetylglucosaminylsaminyltransferase that transfers N-acetylglucosamine from UDP-N-acetylglucosamine to Gal3GalNAc-Mucin to yield Gal3(GlcNAc6)GalNAc-Mucin and a 3-N-acetylglucosaminyl transferase that transfers N-acetylglucosamine from UDP-N-acetylglucosamine to Gal3(GlcNAC6)GalNAc-mucin to yield GlcNAc3Gal3 (GlcNAc6)GalNAc-Mucin were purified from the microsomal fraction of swine trachea epithelium. The 1,6-N-acetylglucosaminyltransferase was purified about 21,800-fold by procedures which included affinity chromatography on DEAE columns containing bound asialo Cowper's gland mucin glycoprotein with Gal1,3GalNAc side chains. The apparent molecular weight estimated by gel filtration was found to be about 60 Kd. The purified enzyme showed a high specificity for Gal1,3GalNAc chains and the most active substrates were mucin glycoproteins containing these chains. The apparent Km of the 6-glucosaminyltrans-ferase for Cowper's gland mucin glycoprotein containing Gal1,3GalNAc chains was 0.53 µM; for UDP-N-acetylglucosamine, 12 µM; and for Gal 1,3GalNAc NO₂ø, 4 mM. The activity of the 6-glucosaminyltransferase was dependent on the extent of glycosylation of the Gal3GalNAc chains in Cowper's gland mucin glycoprotein.The best substrate for the partially purified 3-Glucosaminyltransferase was Cowper's gland mucin glycoprotein containing Gal1,3(GlcNAc6)GalNAc side chains. This enzyme showed little or no activity with intact sialylated Cowper's gland mucin glycoprotein or derivatives of this glycoprotein containing GalNAc or Gal1,3GalNAc side chains.The radioactive oligosaccharides formed by these enzymes in large scale reaction mixtures were released from the mucin glycoproteins by treatment with alkaline borohydride, isolated by gel filtration on Bio-Gel P-6 and characterized by methylation analysis and sequential digestion with exoglycosidases. The oligosaccharide products formed by the 6- and 3-glucosaminyltransferases were shown to be Gal3(GlcNAC6) GalNAc and GlcNAc3 Gal3(GlcNAC6)GalNAc respectively.Taken collectively, these results demonstrate that swine trachea epithelium contains two specific N-acetylglucosaminyltransferases which catalyze the initial branching and elongation reactions involved in the synthesis of O-linked oligosaccharide chains in respiratory mucin glycoproteins. The first enzyme a 6-glucosaminyltransferase converts Gal3GalNAc chains in mucin glycoproteins to Gal3(GlcNAc6)GalNAc chains. This product is the substrate for a second 3-glucosaminyltransferase which converts the Gal3(GlcNAc6)GalNAc chains to GlcNAc3Gal(GlcNAc6)GalNAc chains in the glycoprotein. The 3-glucosaminyltransferase did not utilize Gal3GalNAc chains as a substrate and this results in an ordered sequence of addition of N-acetylglucosamine residues to growing oligosaccharide chains in tracheal mucin glycoproteins.Abbreviations NeuNAc N-acetylneuraminic acid - GalNAcol N-acetylgalactosaminitol - CGMG Cowper's gland mucin glycoprotein - GalNAc-CGMG Cowper's gland mucin glycoprotein containing GalNAc side chains O-glycosidically linked to serine or threonine - Gal3GalNAc-CGMC Cowper's gland mucin glycoprotein containing Gal3GalNAc side chains - MES 2-(N-morpholino) Ethane Sulfonic acid - PBS Phosphate Buffered Saline     

Enrichment and efficient screening of ES cells containing a targeted mutation: the use of DT‐A gene with the polyadenylation signal as a negative selection maker

Gene targeting in embryonic stem (ES) cells via homologous recombination can occur at very low frequency. In order to enrich homologous recombinants before screening, a negative selection marker, such as thymidine kinase (TK) and diphtheria toxin A fragment (DTA), has been commonly used. In this study, we developed a negative selection marker using DTA gene with polyadenylation signal and it was designated DTApA. To determine the difference in targeting efficiency of the negative selections, we constructed three different targeting vectors for each negative selection (first, TK at the 3 end, second, TK at both the 5 and 3 ends <2 X TK>, and third, DTApA at the 5 end of the homologous sequences). Gene targeting experiments using these constructs clearly showed that negative selection using DTApA was more efficient than that using TK for homologous recombination and that negative selection using DTApA was as efficient as that using 2 X TK. Considering the fact that the use of DTApA is more convenient for construction of targeting vectors than that of 2 X TK, DTApA is an efficient negative selection marker.In addition, we examined long and accurate PCR (LAPCR) for screening gene targeted clones. The use of LAPCR with genomic DNAs from ES cell clones facilitated simple detection of homologous recombinants, suggesting that the screening with LAPCR is compatible with the use of longer homologous sequences of both arms in vector design. Our results indicate that the use of DTApA for negative selection together with the application of LAPCR for screening ensures efficient and timesaving screening for homologous recombinants.     

Histoenzymatic characterization of sub-types of Type I fibres in fast muscles of rats

Summary The histochemical activities of succinic dehydrogenase (SDH), myofibrillar Adenosine triphosphatase (ATPase) and alpha glycerophosphate dehydrogenase were studied in serial sections of rat vastus lateralis (red) (RVL), gastrocnemius and diaphragm muscles. Three main fibre-types were distinguished. The Type I fibres of RVL and gastrocnemius muscles fell into two distinct groups: one category-Type IA showed very low ATPase activity. The second category of Type IB fibres displayed moderate ATPase reaction. The Type IA fibres were divisible into two sub-groups when tested for SDH reaction. Type IA₁ fibres possessed a homogenous distribution of diformazan·granules throughout the fibre: Type IA₂ fibres displayed characteristic moth-eaten pattern of diformazan localization. The diaphragm muscle did not show either Type IB or Type IA₂ varieties. The great majority of Type I fibres were sub-type IA₁ in the three fast muscles studied. It is also demonstrated here that an inherent heterogeneity exists between Type I fibres of diaphragm and leg muscles in regard to -GPD localization. This histochemical data emphasizes the fact that subdivision of Type I striated muscle fibres of mammalian animals into two sub-types is only approximate and that a further subcategorization is possible.     

Iron sulfur proteins of the purple sulfur bacterium Ectothiorhodospira shaposhnikovii

In addition to several cytochromes three iron sulfur proteins were detected in mixotrophically grown cells of Ectothiorhodospira shaposhnikovii, a member of the Chromatiaceae. They were identified as a bacterial ferredoxin and two high potential iron sulfur proteins (HIPIPs). The two HIPIPs were purified and characterized. They were named according to their differing retention times on a DEAE-cellulose column using a continuous NaCl gradient: early and late HIPIP. The HIPIPs contain 4 mol of non-heme iron and 4 mol of acid labile sulfur per mol protein. Under the conditions of purification the early HIPIP (E _{m, 7}+270 mV) was present in a semi-reduced state. Using ion-exchange chromatography the early HIPIP could be split into a reduced green-brown (pI=3.7) and an oxidized red-brown (pI=3.9) fraction. The late HIPIP (pI=3.8) showed a midpoint potential of only+155 mV, the lowest redox potential of a HIPIP described so far.Non-common abbreviations HIPIP high potential iron sulfur protein - MOPS 3(N-morpholino)propane sulfonate - SDS sodium dodecylsulfate     

Primary structure of the hemoglobin beta-chain of rose-ringed parakeet (Psittacula krameri)

The primary structure of Rose-ringed Parakeet hemoglobin -chain was established, completing the analysis of this hemoglobin. Comparisons with other avian -chains show variations smaller than those for the corresponding -chains. There are 11 amino acid exchanges in relationship to the only other characterized psittaciform -chain, and a total of 35 positions are affected by differences among all avian -chains analyzed (versus 61 for the -chains). At three positions, the Psittacula -chain has residues unique to this species. Three ₁₁ contacts are modified, by substitutions at positions 51, 116, and 125.     

Ecological theories and Dutch nature conservation

This paper aims to achieve insight into various ecological theories in the Netherlands which have different, and sometimes opposing, views on the conservation of nature. Interviews, publications and archival research brought to light four separate theories: vitalistic/holistic, dynamic, cybernetic and chaos. Diversity is reached through stability according to vitalistic/holistic and cybernetic theories, but through change and instablility according to the dynamic and chaos theories. These two groups are working apart, and continue to have their own ideas. Prediction of the future is only possible with the vitalistic/holistic and cybernetic theories. Ecologists who adhere to these theories feel responsible and able in different ways to change ecological nature towards desirable end goals. The other two theories, dynamic and chaos, appear to be less activist.     

Nonspecific (“pseudo-plasmal”) dye-binding in the Feulgen nuclear stain and its blocking by azocarmin G

Summary In Feulgen nuclear staining nonspecific dye-binding due to the pseudo-plasmal reaction is intensified in isolated cells with intact cytoplasm, and cannot be eliminated by the post-irradiation method. Fluorescence intensity in the cytoplasm sometimes exceeds that of specific nuclear fluorescence, especially in brain and heart muscle cells, and it was almost impossible to perform cytofluorometric DNA quantification on such specimens. Various kinds of aldehyde-blocking agents such as sodium borohydride, 2,4-dinitrophenylhydrazine, aniline, and sodium pyrosulfite were effective in reducing the pseudo-plasmal reaction. But the blocking effects were not complete because of additional release of reactive aldehyde groups during subsequent Feulgen hydrolysis. Acidic azocarmin G produces a complete block of all pseudo-plasmal reaction in acriflavine-Feulgen nuclear staining, allowing accurate DNA-cytofluorometry to be carried out.     

Feedback regulation of RNA polymerase subunit synthesis after the conditional overproduction of RNA polymerase in Escherichia coli

Summary The and subunit of RNA polymerase are thought to be controlled by a translational feedback mechanism regulated by the concentration of RNA polymerase holoenzyme. To study this regulation in vivo, an inducible RNA polymerase overproduction system was developed. This system utilizes plasmids from two incompatibility groups that carry RNA polymerase subunit genes under lac promoter/operator control. When the structural genes encoding the components of core RNA polymerase (, and ) or holoenzyme (, , and ⁷⁰) are present on the plasmids, induction of the lac promoter results in a two fold increase in the concentration of functional RNA polymerase. The induction of RNA polymerase overproduction is characterized by an initial large burst of synthesis followed by a gradual decrease as the concentration of RNA polymerase increases. Overproduction of RNA polymerase in a strain carrying an electrophoretic mobility mutation in the rpoB gene results in the specific repression of synthesis off the chromosome. These results indicate that RNA polymerase feedback regulation controls synthesis in vivo.     

Silver nitrate and aminoethoxyvinylglycine affect <Emphasis Type="Italic">Agrobacterium</Emphasis>-mediated apple transformation

In Fuji, the production of ethylene was increased with the addition of AgNO₃ and inhibited with the addition of 10 M aminoethoxyvinylglycine (AVG). The addition of 80 M AgNO₃ to transformed explants of Fuji cultured on selection medium resulted in increased ethylene production (20 l l^–1) at 3 weeks. Under examining the effect of AgNO₃ in Fuji, the 40 M AgNO₃ showed with higher 33.8% and 6.5% in the efficiency of regeneration and transformation. However, ethylene production in Gala explants treated with 10M AgNO₃ (3 l l^–1) decreased after 2 weeks compared with the control (5 l l^–1). Although the regeneration efficiency of Gala with 10 M AgNO₃ was higher (41.1%) than the control (20.1%), there was no significant difference in the transformation efficiency at the same concentration. Shoot regeneration of Fuji and Gala was completely inhibited with 10 M AVG. These results suggest that the addition of AgNO₃ affects the efficiency of Agrobacterium-mediated gene transfer in Fuji.Eun Soo Seong, Ill Min Chung- These two Authors Contributed equally to this work