A novel epitope of CD59 expressed by primitive human hematopoietic progenitors

A novel epitope of CD59 expressed by primitive human hematopoietic progenitors
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
   Experimental Hematology 29 (2001) 1474–1483 0301-472X/01 $–see front matter. Copyright © 2001 International Society for Experimental Hematology. Published by Elsevier Science Inc. PIIS0301-472X(01)00745-7  A novel epitope of CD59expressed by primitive human hematopoietic progenitors  Paul J. Simmons   d  , Andrew C.W. Zannettino   a  , Dee Harrison-Findik    a  ,Bernadette Swart   a  , Stephen Tomlinson   b  , Beth Hill   c  , and Jeannie A. Javni   d   a   Hanson Centre for Cancer Research, Matthew Roberts Laboratory, Institute of Medical and Veterinary Science, Adelaide, Australia; b   Medical University of South Carolina, Dept. Microbiology and Immunology, Charleston, SC., USA; c  Corixa Corporation, Autoimmune Disease Research, Redwood City, Calif., USA; d  Stem Cell Laboratory, The Peter MacCallum Cancer Institute, East Melbourne, Victoria, Australia  (Received 23 April 2001; revised 13 August 2001; accepted 21 August 2001)  Objective.  The aim of this study was to determine the identity of the cell surface molecule onprimitive hematopoietic cells recognized by monoclonal antibody HCC-1.   Materials and Methods.  Screening of a cDNA expression library prepared from human bonemarrow stromal cells with HCC-1 yielded a single cDNA, which when expressed in FDCP-1cells, resulted in the specific acquisition of HCC-1 binding. The cDNA demonstrated completeidentity with CD59, a phosphoinositol glycan–linked membrane protein that protects cellsagainst autologous complement attack. The ubiquitous expression of CD59 is in marked con-trast to the restricted reactivity of HCC-1. Studies were performed to examine the basis for thenovel specificity of HCC-1 for CD59. The epitope on CD59 identified by HCC-1 was mappedusing a series of rat/human CD59 chimeric proteins. Immunoprecipitation analyses were per-formed to determine whether CD59 associates with other membrane proteins.   Results.  Mutagenesis of Asn18 did not alter the binding of HCC-1 to CD59, suggesting thatN-linked carbohydrates are not responsible for the binding specificity of HCC-1. The epitopefor HCC-1 was shown to differ from that identified by previously described CD59 antibodies,encompassing residues A31, L33, R55, and L59. An 80 kDa protein co-immunoprecipitated with CD59 in the HCC-1     cell line HL-60 but not in HCC-1     K562 cells.  Conclusion.  Collectively, these data support the hypothesis that the unique specificity of HCC-1 for CD59 is due in part to recognition of a novel epitope, which is masked as a result of association with an as yet unidentified 80 kDa protein.© 2001 International Society for Ex- perimental Hematology.Published by Elsevier Science Inc.  A considerable body of evidence demonstrates that in hu-man hematopoietic tissues, hematopoietic stem and progen-itor cells are restricted to a minor subpopulation of mononu-clear cells expressing the sialomucin, CD34 [1–4]. Numerous studies have demonstrated that the CD34     population in fe-tal and adult hematopoietic tissues is heterogeneous, com-prising a hierarchy of closely related yet functionally dis-tinct cell populations [5–10]. Monoclonal antibodies thatsubdivide the CD34     cell population, are therefore of par-ticular interest, since they provide the means to isolate andto analyze the functional properties of defined subpopula-tions of primitive hematopoietic cells. Accordingly, a num-ber of antibody reagents have been described, which iden-tify cell surface glycoproteins, that to a greater or lesserextent are differentially expressed by hierarchically primi-tive cells (including stem cells) and more mature, lineage re-stricted clonogenic progenitors. These include CD33, CD38,and Thy-1/CD90 [11,12; reviewed in 13].This study describes the generation of a murine mAb,HCC-1, which exhibits limited reactivity with both periph-eral blood and BM-derived cells and, more importantly,identifies a subset of the CD34     cell population that func-tionally exhibit multipotentiality in in vitro and in vivomodel systems. In this report we describe the cloning of acDNA corresponding to the cell surface molecule recog-nized by HCC-1. Surprisingly, HCC-1 was found to identifyCD59, an 18–20 kDa phosphoinositol glycan–linked mem-  Offprint requests to: Paul J. Simmons, Ph.D., Stem Cell Research Lab-oratory, Peter MacCallum Cancer Institute, St. Andrew’s Place, East Mel-bourne VIC 3002, Australia; E-mail:   P.J. Simmons et al./Experimental Hematology 29 (2001) 1474–1483  1475  brane protein that protects cells against autologous comple-ment attack [14]. CD59 is uniformly expressed by most nu-cleated cells of the hematopoietic system including all theBM-derived CD34     cells [15,16]. In contrast, the HCC-1–defined epitope of CD59 is absent from the majority of mononuclear cells in peripheral blood and bone marrow andis differentially expressed among CD34     progenitors, withthe highest level present on a subset that is highly enrichedfor pluripotent stem cells [17]. The studies described hereinwere therefore designed to elucidate the physical basis forthe novel specificity of HCC-1 for CD59.  Materials and methods  Generation and characterizationof the HCC-1 monoclonal antibody  Monoclonal antibody (mAb) HCC-1 was generated following im-munization of BALB/c mice with the CD34     cells from adult hu-man marrow. Briefly, BALB/c mice were injected intrasplenicallywith 10   6  CD34     cells and intraperitoneally 2 times with 3   10   6  CD34     cells in 300   L phosphate-buffered saline (PBS) at 3weekly intervals. Ten days after the final intraperitoneal immuni-zation, mice were boosted with the same number of cells adminis-tered intrasplenically. Four days later the spleen cells were fusedwith the NS-1 myeloma cell line using polyethylene glycol (PEG,Boehringer, Mannheim, Germany). The resulting hybridomaswere grown in RPMI 1640 containing 20% (v/v) fetal calf serum(FCS, PA Biologicals, Sydney, Australia) supplemented with hy-poxanthine-aminopterin-thymidine (HAT, Sigma Chemical Co.,St. Louis, MO, USA). Culture supernatants were screened firstlyfor lack of reactivity with peripheral blood mononuclear cells(PBMNC). PBMNC nonreactive supernatants were then screenedby FACS analysis on the immunogen, BM-derived CD34     cells.Hybridomas of interest were cloned thrice by limiting dilution.HCC-1 mAb was isotyped as IgM by means of a mouse mono-clonal antibody isotyping ELISA (Boehringer Mannheim, Mann-heim, Germany), as recommended by the manufacturer.  Preparation of bone marrow mononuclear cells (BMMNC)  Normal bone marrow (BM) was aspirated into preservative-free,sodium heparin–containing tubes (1000 units/mL; Fisons Pharma-ceuticals, Homebush, NSW, Australia), from the sternum and pos-terior iliac crest of healthy young volunteers following informedconsent. The Human Ethics Committee of the Royal AdelaideHospital approved the use of normal BM cells for these studies.Low-density bone marrow mononuclear cells (BMMNC) werecollected after centrifugation at 400  g  over Ficoll-Hypaque (Lym-phoprep, 1.077 g/dL; Nycomed Pharma AS, Oslo, Norway) for 30minutes at room temperature. Mononuclear cells were obtained byselecting the interface cells that were subsequently washed thriceby centrifugation at 4    C in HHF [Hanks Balanced Salt Solution(HBSS); Life Technologies, Gaithersburg, MD, USA] supple-mented with 20 mM HEPES, pH 7.35 and 5% (v/v) FCS.   Isolation of CD34    cells from normal human bone marrow  BMMNC obtained as described above were incubated in blockingbuffer (HHF supplemented with 2% normal human serum) for 30minutes on ice as described above. Labeling was performed withthe anti-CD34 HPCA-2-PE antibody (Becton-Dickinson, Moun-tain View, CA, USA) as previously described. Cell sorting wasperformed using a FACStar   PLUS  cell sorter. The threshold for se-lection of CD34     cells was based on the level of staining obtainedwith an isotype-matched control IgG   1  -PE antibody (Coulter, Hi-aleah, FL, USA). CD34     cells within the lymphocyte/blast regionwere sorted into Iscove’s modification of DMEM (IMDM) supple-mented with 50 Kunitz units/mL DNase I (Sigma) and 20% FCS.Purity of the separated CD34     cells, as assessed by analysis, wasroutinely greater than 98%. Alternatively, CD34     cells were iso-lated by immunomagnetic selection. Briefly, BMMNC obtained asdescribed above were washed twice with ice-cold HHF bufferprior to the addition of anti-CD34 Dynabeads (Dynal, Oslo, Swe-den) at a 1:1 ratio of beads:cells. This suspension was incubated at4    C on a rotary mixer for 60 minutes. Cells rosetted by the CD34Dynabeads were purified by multiple rounds of washing and cap-ture using a magnet (MPC-1, Dynal). The CD34     cells were sub-sequently recovered by incubation of the cell-bead complexes inCD34 Detachabead reagent (Dynal) according to the manufac-turer’s recommendation. The released CD34     cells were washedseveral times in HHF buffer and a portion labeled with HPCA-2-PE (as described above) in order to assess purity. In all experi-ments performed, this procedure yielded populations of CD34     cells that were greater than or equal to 95% pure. CD34 cells wereresuspended in 1    IMDM supplemented with 10% FCS for use inall subsequent assays.  Flow cytometric analysis  Dual-color immunophenotypic analysis was performed as previ-ously described [18]. BMMNC were stained with HPCA-2-FITCor IgG   1  -FITC isotype control (both from Becton-Dickinson,Sunnyvale, CA, USA) in combination with HCC-1 (used as tissueculture supernatant) or the nonbinding IgM isotype control anti-body 1A6 (generously provided by Dr. Leonie Ashman, HCCR,Adelaide). After a 30-minute incubation on ice, the cells werewashed thrice in HHF and incubated for a further 30 minutes in anoptimal dilution of goat anti-mouse IgM-PE (Southern Biotechnol-ogy Associates, Birmingham, AL, USA). After washing as abovein HHF, flow cytometric analysis was performed using a Profile IIflow cytometer (Coulter). 20,000 events were collected per sampleas list mode data and analyzed using Coulter ELITE software.   Immunoprecipitation, PAGE, and Western blotting  Biotinylated NP-40-lysates of cells were prepared as described[19]. Goat anti-mouse Ig–coupled Sepharose (AH-Sepharose 4B,Pharmacia, Piscataway, NJ, USA), was washed twice in TSE (50mM Tris-HCl, 150 mM NaCl, 1 mM EDTA) with 1% (v/v) NP-40and incubated with 400   L of hybridoma supernatant at 4    C for 6hours. The resulting prearmed Sepharose was washed twice in 1%NP-40-TSE and incubated overnight, at 4    C, with cellular lysatefrom 1   10   7  cells. The immunoprecipitates were washed twice in1% (v/v) NP-40-TSE, once in 0.1% (v/v) NP-40-TSE, followed byonce in TSE, pH 8.0. Immunoprecipitates were subsequentlyboiled for 3 minutes in 25   L reducing sample buffer (62.5 mMTris, 3% (w/v) SDS, 10% (v/v) glycerol, and 5% (v/v) 2-mercapto-ethanol) prior to electrophoresis through a 10% (w/v) SDS-poly-acrylamide gel. The proteins were transferred onto Hybond-Cmembrane (Amersham Int., Amersham, Bucks, UK) using a semi-dry blotting apparatus (Hoefer Scientific Instruments, San Fran-cisco, CA, USA). The membrane was blocked by overnight incu-   1476  P.J. Simmons et al./Experimental Hematology 29 (2001) 1474–1483  bation in PBS/3% (w/v) bovine serum albumin (BSA) at roomtemperature, washed four times in PBS/0.5% (v/v) Tween-20, andsubsequently incubated with streptavidin-biotin-HRPO complex(Amersham). The membrane was washed four times in PBS/0.5%(v/v) Tween-20 and immunoreactive proteins were visualized byenhanced chemiluminescence (ECL, Amersham) as recommendedby the manufacturer.   Immunoprecipitation with HCC-1.  FDCP-1 cells expressing theHCC-1 cDNA were washed thrice and lysed at a concentration of 5   10   7  cells/mL in lysis buffer containing protease inhibitors (asabove). The lysate was precleared using goat anti-mouse IgM aga-rose beads (Sigma) for 1 hour at 4    C with rotation, the lysate splitinto two equal aliquots, and immunoprecipitations performed for 2hours at 4    C using the anti-mouse IgM agarose beads prearmedwith either the nonbinding IgM isotype control 1A6.12 or HCC-1.The immunoprecipitates were subsequently washed twice in lysisbuffer prior to the addition of 30   L of nonreducing sample buffer,boiled for 5 minutes, separated on a 15% SDS-polyacrylamide gel,and transferred onto nitrocellulose. The membrane was blocked in5% skim milk in PBS overnight at 4    C and probed with YTH53.1,a rat IgG2a Mab to human CD59, or with 2.4G2 (an isotype-matched rat antibody nonreactive with human cells; BD-Pharmin-gen), both diluted to 2   g/mL in 3% BSA-PBS. The membranewas subsequently washed thrice for 5 minutes per wash with 0.1%Tween-20 in PBS prior to sequential incubation with goat anti-ratbiotin (Jackson Laboratories) at 1:10,000 and streptavidin-HRPO(Amersham) at 1:2000, both for 30 minutes at room temperature.Peroxidase activity was revealed by ECL, as above.  O-glycosidase treatment of lysates.  FDCP1/HCC-1 cells (1.5   10   7  ) were washed thrice in ice-cold PBS and resuspended in lysisbuffer comprising 100 mM sodium acetate pH 5.0/1% NP-40 andprotease inhibitors. After lysis, nuclei were pelleted and lysateswere divided into three aliquots. To one aliquot an equal volume of nonreduced sample buffer was added, the sample boiled and storedat   70    C until ready for use. The second aliquot received 4   L of sodium acetate buffer and the third aliquot 4   L (2 mU) of O-gly-cosidase (Roche Diagnostics Australia Pty. Ltd.). Aliquots two andthree were incubated at room temperature for 30 hours prior to theaddition of an equal volume of nonreducing sample buffer. Sam-ples were boiled and stored at   70    C. Electrophoresis and transferwere performed as described above and blots were blocked in 5%skim milk–PBS/Tween 20 prior to probing with the control mono-clonal antibody 1A6.12 or HCC-1 and thereafter sequentially withbiotinylated goat anti-mouse IgM (Southern Biotechnology Asso-ciates, Birmingham, AL, USA) and ABC reagent (Vector Labora-tories, Burlingame, CA, USA). Blots were washed thrice in PBSTbetween incubations and peroxidase activity revealed by ECL, asabove.   Expression cloning of the HCC-1 cDNA  cDNA synthesized from human bone marrow stromal cell–derivedmRNA was used to construct a cDNA library in the retroviral vec-tor pRUF  neo  as previously described [20]. Following transfectionof the amphotropic packaging line PA317, virus-containing super-natant was used to infect the ecotropic packaging cell line,   2. Fol-lowing selection of G418-resistant   2 cells, retroviral supernatantwas harvested and used to infect the murine factor–dependent cellline FDCP-1. Infected cells were selected for G418 resistance andcells expressing the cDNA encoding the cell surface moleculeidentified by HCC-1 were selected using the monoclonal antibodyand expanded into clonal populations using multiple rounds of im-munomagnetic bead (Dynabead) selection followed by FACS sort-ing as previously described [20]. Genomic DNA prepared fromFDCP-1 cells was used in a polymerase chain reaction (PCR) usingretroviral-specific primers [20] to recover proviral cDNA inserts.cDNA clones generated by PCR were gel purified and subclonedinto the pGEM T vector (Promega, Madison, WI, USA) as recom-mended by the manufacturer. 500–600 base pair (bp) 5    and 3    partial sequence data obtained from the clones were analyzed byaccessing the Genbank and European Molecular Biology labora-tory (EMBL) data bases at the National Centre for Biotechnologi-cal Information (NCBI).   In vitro mutagenesis of CD59 cDNA  The in vitro mutagenesis was carried out either by the AlteredSites in vitro mutagenesis system (Promega) or by the Quick-Change (Strategene, La Jolla, CA, USA) procedure as recom-mended by the manufacturers. Utilizing the Altered Sites in vitromutagenesis system, a 651 bp Sma  I and Sph  I restriction fragmentof the CD59 cDNA, derived by PCR, was ligated into the 5680 bppALTER vector. The mutagenesis reaction involved the annealingof the ampicillin repair oligonucleotide and the MUT59 mutagenicoligonucleotide to the recombinant pALTER single-stranded DNAfollowed by synthesis of the mutant strand with T4 DNA polymer-ase. Mutant CD59 and WT constructs were distinguished by theincorporation of a  Bam  HI restriction endonuclease site in the se-quence of the MUT59 mutagenesis oligonucleotide and confirmedby automated sequence analysis utilizing the   21M13 and M13priming sites present in the pALTER vector. The FDCP-1 cell linewas transduced with WT and mutant isoforms of CD59 DNA as de-scribed above. Protein expression was assessed by indirect immuno-fluorescence and flow cytometry.  Results  We have previously described the properties of a mono-clonal antibody, HCC-1, that was produced following im-munization with human bone marrow–derived CD34     cells.In accordance with screens that were employed to identifyHCC-1, the antibody exhibits minimal reactivity withmononuclear cells in the peripheral blood and bone marrowand identifies a subpopulation of CD34     cells in BM. TheCD34     HCC-1     population includes candidate hematopoie-tic stem cells as demonstrated by multilineage reconstitu-tion of human hematopoiesis in SCID-hu mice [17]. In addi-tion, HCC-1 also binds to cultures of bone marrow–derivedstromal cells (data not shown). We therefore used the HCC-1antibody to screen a cDNA expression library preparedfrom human BM stromal cells in the retroviral vectorpRUFneo, as previously described [20]. A single cDNA of 1.2 kb was identified that, when reintroduced by retroviralinfection into the murine myeloid cell line FDCP-1, specifi-cally conferred an HCC-1 binding phenotype to the cells(Fig. 1A). Sequence analysis of HCC-1 cDNA revealedcomplete identity with CD59 (data not shown), a GPI-linked 18–20 kDa cell surface protein. Two monoclonal an-tibodies with well-defined specificity for CD59, MEM-43   P.J. Simmons et al./Experimental Hematology 29 (2001) 1474–1483  1477  [21] and YTH 53.1 [22], both demonstrated specific bindingto the FDCP-1 cells expressing the HCC-1 cDNA (Fig. 1Band C). Similar reactivity was also seen with a third CD59-reactive antibody, 2/24 [23](data not shown). Furthermore,immunoprecipitation of FDCP-1–HCC-1 cells with HCC-1antibody followed by immunoblotting with YTH53.1 spe-cifically identified a 16 kDa band (Fig. 1D), consistent withthe molecular weight of human CD59 [21,22,24].Previous work relating to CD59 has demonstrated ubiq-uitous expression of this molecule within the human he-matopoietic system, including not only primitive hemato-poietic stem and progenitor cells but also their matureprogeny including all peripheral blood leukocytes, platelets,and erythrocytes [15,16]. This pattern of expression of CD59 is in marked contrast to that demonstrated by HCC-1,most strikingly demonstrated by the contrasting pattern of reactivity of MEM-43 and HCC-1 on BMMNCs and PB-MNCs (Fig. 2A–C). While the mAb MEM-43 binds to 85%of BMMNCs, HCC-1 reacts with only 20% (range 20–50%,n   10) of cells (Fig. 2A and B) in accordance with previousobservations [17]. Similarly, on PBMNCs, MEM-43 exhib-its binding at high levels to virtually all cells, while HCC-1binds at low levels to approximately 3% (range 3–9.8%, n   8) (Fig. 2C). This difference in antibody reactivity is also seen on CD34     BMMNCs: MEM-43 binds to essentially allCD34     cells while HCC-1 reacts with a subpopulation com-prising approximately 30% (range 30–70%, n   7) of theCD34     population in accordance with previous observa-tions [17]. The differential expression of the HCC-1 epitopewas also evident on hematopoietic cell lines. K562, HEL-DR, and KG1a cells demonstrated reactivity with mAbs 2/ 24 and MEM-43 and with HCC-1. In contrast, HL-60 cells,despite uniform binding of 2/24 and MEM-43, failed to re-act with HCC-1 (data not shown). Collectively, these datademonstrate that HCC-1 exhibits a unique specificity forCD59. Subsequent experiments were therefore performed toinvestigate the molecular basis for the differential expres-sion of the HCC-1 epitope.Glycosylation of CD59 has been shown to play a role inthe complement inhibitory function of CD59 [25,26]. Totest whether the differential binding of HCC-1 was depen-dent upon a carbohydrate epitope, the solitary predicted siteof N-glycosylation, Asn   18  , was mutated by in vitro mutagen-esis. FDCP-1 cells expressing the mutant and wild-typeCD59 molecules were immunolabeled with both the HCC-1and 2/24 antibodies. This analysis demonstrated comparablebinding of HCC-1 to wild-type and mutant CD59-express-ing cells, suggesting that HCC-1 binding was not dependent Figure 1. The cloned HCC-1 cDNA confers binding of HCC-1 mAb to retrovirally infected FDCP-1 cells and HCC-1 identifies the product of the CD59gene. A 2.1 kb NotI-NotI restriction fragment of the HCC-1 (CD59) cDNA (harboring both the entire coding sequence and the 5   and 3   noncoding regions)was subcloned into the pRUF neo  vector and subsequently introduced into FDCP-1 cells by retroviral transduction. The resultant G418-resistant cell popula-tion was stained with (A)  HCC-1, (B)  MEM-43, and (C)  YTH53.1 antibodies (dark lines) and analyzed by flow cytometry. IgM, IgG2a, and IgG2b nonbind-ing isotype antibodies were used as controls (light broken lines). Data are displayed as single-parameter fluorescence (FITCPE) histograms of 2   10 4  lightscatter gated events, collected as list mode data. (D)  HCC-1 identifies CD59. FDCP-1 cells expressing the HCC-1 (CD59) cDNA were immunoprecipitatedwith HCC-1 or the isotype control IgM antibody 1A6.12. Immunoblots were then probed with the human CD59-specific antibody YTH53.1 or isotype-matched control antibody. YTH53.1 specifically identifies a band of 16 kDa in immunoprecipitates with HCC-1.   1478  P.J. Simmons et al./Experimental Hematology 29 (2001) 1474–1483  on N-linked glycosylation (data not shown). To investigatethe potential contribution of O-linked glycans to HCC-1binding, cell lysates prepared from HCC-1–FDCP-1 weretreated with O-glycosidase to remove O-linked carbohy-drates according to methodologies previously used to re-move O-linked carbohydrate from the sialomucin CD164[27]. Western blotting of these lysates with HCC-1 (Fig. 3)did not reveal any significant difference in antibody bindingfollowing enzyme treatment, suggesting that HCC-1 bind-ing is also not dependent on O-linked glycosylation.Alternatively, the unique specificity of the HCC-1 anti-body may be attributed to the antibody recognition of anepitope on CD59 distinct from that previously defined forother anti-CD59–reactive antibodies [28,29]. To investigatethis possibility, we mapped the epitope for HCC-1 by mu-tating selected amino acid residues on human CD59 to thecorresponding residues of rat CD59, a strategy we antici-pated would avoid introducing major conformational and/orstructural changes into the human CD59 molecule. Usingthis approach, the epitope for HCC-1 was found to map tothe residues A31, L33, R55, and L59 on CD59 (Fig. 4 andTable 1). The three-dimensional protein structure of theCD59 molecule [30,31] has revealed that the residue L33 isat an exposed surface loop and residue R55 resides in the    -helix while residues A31 and L59 are located at the endsof the central three-stranded antiparallel   -sheet.The residues forming the putative HCC-1 epitope residetoward the rear of the molecule closer to the plasma mem- Figure 2. Expression of HCC-1 and MEM-43 reactive epitopes by bone marrow and peripheral blood mononuclear cells. Dual-parameter immunofluores-cence of BMMNCs labeled with both HPCA-2 (anti-CD34) and HCC-1 mAb (A)  or MEM-43 mAb (B)  was carried out as described in Materials and meth-ods. Each two-color plot was generated with 2   10 4  light-scatter gated events, collected as list mode data. (C)  Single-parameter fluorescence histogram pro-files of PBMNCs labeled with HCC-1 mAb (dark line) or MEM-43 mAb (broken line). IgG2a and IgM (light lines) nonbinding isotype antibodies were usedas control. Figure 3. The binding of mAb HCC-1 is not dependent on O-linked gly-cosylation. A cell lysate was prepared from FDCP1/HCC-1 cells anddivided into three aliquots. To one aliquot (A)  nonreduced sample bufferwas added. The second aliquot (B)  was mock treated with sodium acetatebuffer and the third aliquot (C)  was treated with O-glycosidase. Aliquotstwo and three were incubated at room temperature for 30 hours.
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks