<p><b>PAPER 1</b></p><p>In the first paper we show that reducing the average molecular weight from ~350 kDa to <6kDa by acid hydrolysis diminished the cell-stimulating activity of poly-M, measured as TNFproduction from human monocytes. However, the activity of the resulting oligomers (M-blocks) was greatly enhanced when covalently attached to particles (plastic beads or biodegradable albumin particles). Similar results were obtained with detoxified/deacylated LPS (DLPS) and glucuronic acid polymers (C6OXY), but not with G-blocks that by themselves are not active. These results suggest that the supramolecular structure affects the potency of polysaccharide stimuli, and that M-blocks attached to biodegradable albumin particles could possibly be exploited as an immunostimulant for protection against various diseases.</p><p><b>PAPER 2</b></p><p>In paper 2, according to the reviewers suggestion, the designation M-polymers of different molecular size was used in place of poly-M (~350 kDa) and M-blocks (~3 kDa). In this study we demonstrated that M-blocks and DLPS attached to particles engaged different receptors than soluble poly-M and DLPS in activation of monocytes. By using blocking mAbs to CD14, CD11b and CD18, we found that particulate stimuli employed the β2- integrin CD11b/CD18 in addition to the shared CD14 for signaling TNF-production. Moreover, whereas poly-M only bound to CD14-expressing CHO-cells, M-particles preferentially bound to CHO-cells expressing β2-integrins. However, the DLPS- and M-particles failed to activate NF-κB-translocation in CHO-cells co-transfected with CD14 and β2-integrins, suggesting that additional molecules are required for activation of CHO-cells. The major conclusion drawn from this work is that the supramolecular structure, in addition to influence the potency, affects the cellular receptor engagement by carbohydrates like poly-M and DLPS. This points to the importance of comparing the mechanisms involved in activation of immune cells by soluble bacterial components and whole bacteria to achieve a better understanding of inflammatory diseases like sepsis.</p><p><b>PAPER 3</b></p><p>Poly-M activates cells in a CD14-dependent manner, but CD14 is linked to the membrane with a GPI-anchor and mediates activation by interaction with other, signal-transducing molecules, like the TLRs. By using blocking mAbs to TLR2 (generated in our lab, paper 5) and TLR4, we found that both receptors were involved in mediating TNF-production from human monocytes in response to poly-M. Furthermore, TLR4 mutant (C3H/HeJ) and knockout (TLR4-/-) murine macrophages were completely non-responsive to poly-M, whereas TLR2-deficient macrophages showed reduced TNF-responses. These findings indicate that CD14, TLR2 and TLR4 on primary cells all participate in cytokine-induction by poly-M, and that TLR4 may be necessary for activation.</p><p><b>PAPER 4</b></p><p>In addition to CD14, β2-integrins have been implicated in LPS-induced cellular activation, and in this study we compared the involvement of CD14 and β2-integrins in TNF-production and NF-κB-activation induced by LPS and GBS cell wall fragments. With blocking mAbs to CD14 and CD18 we found that LPS and GBS cell walls shared CD14, but in addition the cell walls employed CD11/CD18 in mediating TNF-production from human monocytes. Both stimuli specifically induced NF-κB-translocation in CD14-transfected CHO-cells, but only LPS could activate cells transfected with CD11/CD18. The lack of response to GBS cell walls in CD11/CD18-transfected CHO-cells indicated that the cell walls need CD14 for cell activation. Further in paper 4 we demonstrate the ability of GBS cell walls to activate LPS-hyporesponsiv C3H/HeJ mouse macrophages, suggesting that LPS and GBS cell walls employ different receptors/signaling mechanisms in murine macrophages.</p><p><b>PAPER 5</b></p><p>When it was discovered that human TLR2 and TLR4 are involved in microbial recognition, we started to generate a mouse mAb to human TLR2, and in paper 5 we report the production and characterization of the mAb TL2.1. We subsequently used this mAb to evaluate the role of TLR2 in mediating activation by heat-killed GBS and <i> L monocytogenes</i>.<i> L. monocytogenes</i>, but not GBS, activated TLR2-transfected CHO-cells to IL-6-production, and the response was inhibited by TL2.1. A CD14 mAb and TL2.1 both inhibited TNF-production from monocytes induced by <i>L. monocytogenes</i>, but neither mAb affected the TNF-response triggered by GBS. Our results suggest that CD14 and TLR2 are engaged in cell activation by <i>L. monocytogenes</i>, but that neither receptor seem to be involved in activation by GBS. This study was the first to show that human TLR2 can discriminate between two G+ bacteria.</p><p><b>PAPER 6</b></p><p>In paper 6 we report the generation of a new TLR2 mAb, TL2.3, that stained with the same specificity as TL2.1 (anti-TLR2, paper 5). We used these mAbs to investigate the expression of TLR2 protein in human cells. We found that TLR2 was highly expressed in blood monocytes, less in granulocytes, and not present in lymphocytes. The protein level was measured on quiescent and activated cells by extra- and intracellular flow cytometry, and by immunoprecipitation of TLR2 from metabolic S35-labeled cells. Surprisingly, TLR2 protein was detected in activated B-cells located in lymphoid germinal centers, indicating that subsets of lymphocytes may express TLR2. We further show that TLR2 protein was differentially regulated on monocytes and granulocytes after exposure to LPS, pro- or anti-inflammatory cytokines. However, we could not correlate the regulation of TLR2 to cellular responses, as for instance the three anti-inflammatory cytokines TGFβ, IL-4 and IL-10 all inhibited lipopeptideinduced TNF-production, but either did not affect, reduced, or increased the level of surface TLR2, respectively. Thus, the biological significance of TLR2-regulation remains to be found.</p>

<b>PAPER 1</b> In the first paper we show that reducing the average molecular weight from ~350 kDa to &lt;6kDa by acid hydrolysis diminished the cell-stimulating activity of poly-M, measured as TNFproduction from human monocytes. However, the activity of the resulting oligomers (M-blocks) was greatly enhanced when covalently attached to particles (plastic beads or biodegradable albumin particles). Similar results were obtained with detoxified/deacylated LPS (DLPS) and glucuronic acid polymers (C6OXY), but not with G-blocks that by themselves are not active. These results suggest that the supramolecular structure affects the potency of polysaccharide stimuli, and that M-blocks attached to biodegradable albumin particles could possibly be exploited as an immunostimulant for protection against various diseases. <b>PAPER 2</b> In paper 2, according to the reviewers suggestion, the designation M-polymers of different molecular size was used in place of poly-M (~350 kDa) and M-blocks (~3 kDa). In this study we demonstrated that M-blocks and DLPS attached to particles engaged different receptors than soluble poly-M and DLPS in activation of monocytes. By using blocking mAbs to CD14, CD11b and CD18, we found that particulate stimuli employed the β2- integrin CD11b/CD18 in addition to the shared CD14 for signaling TNF-production. Moreover, whereas poly-M only bound to CD14-expressing CHO-cells, M-particles preferentially bound to CHO-cells expressing β2-integrins. However, the DLPS- and M-particles failed to activate NF-κB-translocation in CHO-cells co-transfected with CD14 and β2-integrins, suggesting that additional molecules are required for activation of CHO-cells. The major conclusion drawn from this work is that the supramolecular structure, in addition to influence the potency, affects the cellular receptor engagement by carbohydrates like poly-M and DLPS. This points to the importance of comparing the mechanisms involved in activation of immune cells by soluble bacterial components and whole bacteria to achieve a better understanding of inflammatory diseases like sepsis. <b>PAPER 3</b> Poly-M activates cells in a CD14-dependent manner, but CD14 is linked to the membrane with a GPI-anchor and mediates activation by interaction with other, signal-transducing molecules, like the TLRs. By using blocking mAbs to TLR2 (generated in our lab, paper 5) and TLR4, we found that both receptors were involved in mediating TNF-production from human monocytes in response to poly-M. Furthermore, TLR4 mutant (C3H/HeJ) and knockout (TLR4-/-) murine macrophages were completely non-responsive to poly-M, whereas TLR2-deficient macrophages showed reduced TNF-responses. These findings indicate that CD14, TLR2 and TLR4 on primary cells all participate in cytokine-induction by poly-M, and that TLR4 may be necessary for activation. <b>PAPER 4</b> In addition to CD14, β2-integrins have been implicated in LPS-induced cellular activation, and in this study we compared the involvement of CD14 and β2-integrins in TNF-production and NF-κB-activation induced by LPS and GBS cell wall fragments. With blocking mAbs to CD14 and CD18 we found that LPS and GBS cell walls shared CD14, but in addition the cell walls employed CD11/CD18 in mediating TNF-production from human monocytes. Both stimuli specifically induced NF-κB-translocation in CD14-transfected CHO-cells, but only LPS could activate cells transfected with CD11/CD18. The lack of response to GBS cell walls in CD11/CD18-transfected CHO-cells indicated that the cell walls need CD14 for cell activation. Further in paper 4 we demonstrate the ability of GBS cell walls to activate LPS-hyporesponsiv C3H/HeJ mouse macrophages, suggesting that LPS and GBS cell walls employ different receptors/signaling mechanisms in murine macrophages. <b>PAPER 5</b> When it was discovered that human TLR2 and TLR4 are involved in microbial recognition, we started to generate a mouse mAb to human TLR2, and in paper 5 we report the production and characterization of the mAb TL2.1. We subsequently used this mAb to evaluate the role of TLR2 in mediating activation by heat-killed GBS and L monocytogenes. L. monocytogenes, but not GBS, activated TLR2-transfected CHO-cells to IL-6-production, and the response was inhibited by TL2.1. A CD14 mAb and TL2.1 both inhibited TNF-production from monocytes induced by L. monocytogenes, but neither mAb affected the TNF-response triggered by GBS. Our results suggest that CD14 and TLR2 are engaged in cell activation by L. monocytogenes, but that neither receptor seem to be involved in activation by GBS. This study was the first to show that human TLR2 can discriminate between two G+ bacteria. <b>PAPER 6</b> In paper 6 we report the generation of a new TLR2 mAb, TL2.3, that stained with the same specificity as TL2.1 (anti-TLR2, paper 5). We used these mAbs to investigate the expression of TLR2 protein in human cells. We found that TLR2 was highly expressed in blood monocytes, less in granulocytes, and not present in lymphocytes. The protein level was measured on quiescent and activated cells by extra- and intracellular flow cytometry, and by immunoprecipitation of TLR2 from metabolic S35-labeled cells. Surprisingly, TLR2 protein was detected in activated B-cells located in lymphoid germinal centers, indicating that subsets of lymphocytes may express TLR2. We further show that TLR2 protein was differentially regulated on monocytes and granulocytes after exposure to LPS, pro- or anti-inflammatory cytokines. However, we could not correlate the regulation of TLR2 to cellular responses, as for instance the three anti-inflammatory cytokines TGFβ, IL-4 and IL-10 all inhibited lipopeptideinduced TNF-production, but either did not affect, reduced, or increased the level of surface TLR2, respectively. Thus, the biological significance of TLR2-regulation remains to be found.

<p>A question of great interest in evolutionary biology is how and why the eukaryotic cell evolved. Several hypotheses have been proposed, ranging from an early emergence of a primitive eukaryotic cell, to various fusion models like the hydrogen hypothesis. Within this context, relevant bacterial gene families and genomes are examined in this thesis.</p><p>The mitochondrion, the energy producing organelle in the eukaryotic cell, is generally believed to be of α-proteobacterial descent. To learn more about mitochondrial evolution, and therefore eukaryotic evolution, the genomes of the α-proteobacteria <i>Bartonella henselae</i> and <i>Bartonella quintana</i> were sequenced. Software was developed and used in the annotation of these genomes.</p><p>Several gene products of nuclear-encoded genes are exported to the mitochondrion. Many of these genes are thought to originate from the emerging organelle. An analysis of the more than 400 genes encoding proteins targeted to the yeast mitochondrion indicates that one set of genes originated from the bacterial symbiont, while the eukaryotic host contributed another. Thus, the mitochondrial proteome has a dual origin.</p><p>The hydrogen hypothesis postulates that the glycolytic genes belong to the group of genes that were transferred from symbiont to host. These genes are thoroughly analysed from a phylogenetic perspective. Contrary to the predictions of the hydrogen hypothesis, the results provide no support for a close relationship between nuclear genes encoding glycolytic enzymes and their α-proteobacterial homologs. </p><p>In general, it is thought that intensive gene transfers may limit our ability to reconstruct gene and species evolution, especially among microbes. A phylogenetic analysis of a large cohort of genes from the AT-rich genome of the γ-proteobacterium <i>Buchnera aphidicola</i> (Sg) resulted in a high fraction of atypical tree topologies, previously interpreted as horizontal gene transfers. By applying methods that accommodate for asymmetric nucleotide substitutions, it is shown that many well-supported gene topologies are drastically altered, so that they now agree with the rRNA topology. The conclusion is that atypical topologies may not necessarily be evidence for horizontal gene transfers. </p>

<p>Cellulose is the most abundant organic compound on earth. A wide range of highly specialized microorganisms, have evolved that utilize cellulose as carbon and energy source. Enzymes called cellulases, produced by these cellulolytic organisms, perform the major part of cellulose degradation.</p><p>In this study the three-dimensional structure of four homologous glycoside hydrolase family 12 cellulases will be presented, three fungal enzymes; <i>Humicola grisea</i> Cel12A, <i>Hypocrea schweinitzii </i>Cel12A, <i>Trichoderma reesei</i> Cel12A, and one bacterial; <i>Streptomyces sp. 11AG8</i> Cel12A. The structural and biochemical information gathered from these and 15 other GH family 12 homologues has been used for the design of variants of these enzymes. These variants have biochemically been characterized, and thereby the positions and the types of mutations have been identified responsible for the biochemical differences between the homologous enzymes, e.g., thermal stability and activity. The three-dimensional structures of two <i>T. reesei</i> Cel12A variants, where the mutations have significant impact on the stability or the activity of the enzyme have been determined. Four ligand complex structures of the WT <i>H. grisea</i> Cel12A enzyme, that have made it possible to characterize the interactions between substrate and enzyme, have also been determined. </p><p>The structural and biochemical studies of these closely related GH family 12 enzymes, and their variants, have provided insight on how specific residues contribute to protein thermal stability and enzyme activity. This knowledge can in the future serve as a structural toolbox, i.e., to design Cel12A enzymes with specific properties and features by introducing subtle changes in structural components of the enzymes. These can then be utilized to develop new industrial products or fine-tune enzymes in already existing applications.</p>

<p>Cyanobacteria are a heterogeneous group of phototrophic microorganisms. Many cyanobacteria have the capacity to fix atmospheric nitrogen. During the process of nitrogen fixation, molecular hydrogen is produced. Three enzymes are directly involved in hydrogen metabolism in cyanobacteria. A nitrogenase, evolving hydrogen during nitrogen-fixation, an uptake hydrogenase, recycling the hydrogen produced by nitrogenase, and a bidirectional hydrogenase that has the capacity to both take up and produce hydrogen. The main objective in this thesis was to examine the transcriptional regulation of both the uptake and the bidirectional hydrogenase in filamentous cyanobacteria.</p><p>The transcriptional regulation of the uptake hydrogenase was demonstrated to be influenced by external conditions in <i>Nostoc muscorum</i> and <i>Nostoc punctiforme</i>. Nickel, molecular hydrogen, and anaerobic conditions all induced the relative amount of uptake hydrogenase transcript. In addition, a transcript could be detected in nitrogen-fixing, but not in non-nitrogen fixing conditions.</p><p>The transcriptional regulation of the bidirectional hydrogenase in <i>N. muscorum</i> and <i>Anabaena</i> PCC 7120 was also examined. The relative amount of transcript from the bidirectional hydrogenase in both strains was demonstrated to increase during anaerobic conditions. Moreover, experiments using <i>N. muscorum</i> demonstrated that addition of nickel also increase the amount of transcript. However, no change in the relative amount of transcript from the bidirectional hydrogenase could be observed by additional hydrogen or during a shift from non-nitrogen fixing to nitrogen fixing conditions.</p><p>The genes responsible for maturation of the hydrogenase were identified, cloned and sequenced in <i>N. punctiforme</i>. The transcription of the genes was examined and all genes were located on a single transcript. Like the uptake hydrogenase, a transcript could be detected under nitrogen-fixing but not under non-nitrogen fixing conditions. </p><p>Initial studies, using microarrays, were used to analyse and compare the transcription of a large set of <i>Anabaena</i> PCC 7120 genes under non-nitrogen and nitrogen-fixing conditions. Both up- and down-regulated genes could be identified.</p><p>This thesis advances the knowledge about the transcriptional regulation of the hydrogenases in filamentous cyanobacteria and can be used as a platform for further experiments aiming at a modified hydrogen metabolism.</p>

<p>Heparan sulfate (HS) is a long unbranched polysaccharide composed of alternating glucosamine and hexuronic (glucuronic or iduronic) acid residues. Modification by sulfate groups in various positions generates a highly heterogeneous molecule. HS is synthesized as a proteoglycan by virtually all cells, and play pivotal functions in signaling and developmental patterning, but also in pathogenic events such as tumor metastasis and microbial adhesion.</p><p>This thesis deals with the properties of enzymes involved in HS chain elongation. Polymerization of the HS chain is believed to be catalyzed by the EXT family of proteins. In humans, the EXT family consists of five members: EXT1, EXT2, EXTL1, EXTL2 and EXTL3; their respective functions in HS biosynthesis are not fully understood. </p><p>In this study, for the first time, successful in vitro HS polymerization on oligosaccharide acceptor substrates was demonstrated, using recombinant EXT1 and EXT1/EXT2 complex. EXT1 formed longer chains than EXT1/EXT2 and their mechanisms of sugar incorporation were different. </p><p>Suppression of EXT1 or EXT2 expression by siRNA in a human cell line resulted in reduction of HS chain length. In contrast, cells transfected with EXTL3 siRNA produced longer HS chains. Overexpression of soluble EXT1, alone or co-expressed with EXT2, resulted in increased chain length, whereas overexpression of soluble EXT2 or EXTL3 has no detectable effect on HS chain elongation. </p><p>Structural analysis of HS from fibroblasts isolated from mice with a hypomorphic mutation in Ext1 showed that they produced significantly shorter HS chains then the wild-type fibroblasts (20 and 70 kDa, respectively). The disaccharide composition of the HS produced by the mutant cells was virtually indistinguishable from that of the wild-type HS, however, the mutant HS chains contained higher proportions of unmodified regions. Mutant cells responded less efficiently than wild-type cells to low concentrations of FGF2, as analyzed by ERK phosphorylation assay.</p>

A question of great interest in evolutionary biology is how and why the eukaryotic cell evolved. Several hypotheses have been proposed, ranging from an early emergence of a primitive eukaryotic cell, to various fusion models like the hydrogen hypothesis. Within this context, relevant bacterial gene families and genomes are examined in this thesis. The mitochondrion, the energy producing organelle in the eukaryotic cell, is generally believed to be of α-proteobacterial descent. To learn more about mitochondrial evolution, and therefore eukaryotic evolution, the genomes of the α-proteobacteria Bartonella henselae and Bartonella quintana were sequenced. Software was developed and used in the annotation of these genomes. Several gene products of nuclear-encoded genes are exported to the mitochondrion. Many of these genes are thought to originate from the emerging organelle. An analysis of the more than 400 genes encoding proteins targeted to the yeast mitochondrion indicates that one set of genes originated from the bacterial symbiont, while the eukaryotic host contributed another. Thus, the mitochondrial proteome has a dual origin. The hydrogen hypothesis postulates that the glycolytic genes belong to the group of genes that were transferred from symbiont to host. These genes are thoroughly analysed from a phylogenetic perspective. Contrary to the predictions of the hydrogen hypothesis, the results provide no support for a close relationship between nuclear genes encoding glycolytic enzymes and their α-proteobacterial homologs. In general, it is thought that intensive gene transfers may limit our ability to reconstruct gene and species evolution, especially among microbes. A phylogenetic analysis of a large cohort of genes from the AT-rich genome of the γ-proteobacterium Buchnera aphidicola (Sg) resulted in a high fraction of atypical tree topologies, previously interpreted as horizontal gene transfers. By applying methods that accommodate for asymmetric nucleotide substitutions, it is shown that many well-supported gene topologies are drastically altered, so that they now agree with the rRNA topology. The conclusion is that atypical topologies may not necessarily be evidence for horizontal gene transfers.

Cellulose is the most abundant organic compound on earth. A wide range of highly specialized microorganisms, have evolved that utilize cellulose as carbon and energy source. Enzymes called cellulases, produced by these cellulolytic organisms, perform the major part of cellulose degradation. In this study the three-dimensional structure of four homologous glycoside hydrolase family 12 cellulases will be presented, three fungal enzymes; Humicola grisea Cel12A, Hypocrea schweinitzii Cel12A, Trichoderma reesei Cel12A, and one bacterial; Streptomyces sp. 11AG8 Cel12A. The structural and biochemical information gathered from these and 15 other GH family 12 homologues has been used for the design of variants of these enzymes. These variants have biochemically been characterized, and thereby the positions and the types of mutations have been identified responsible for the biochemical differences between the homologous enzymes, e.g., thermal stability and activity. The three-dimensional structures of two T. reesei Cel12A variants, where the mutations have significant impact on the stability or the activity of the enzyme have been determined. Four ligand complex structures of the WT H. grisea Cel12A enzyme, that have made it possible to characterize the interactions between substrate and enzyme, have also been determined. The structural and biochemical studies of these closely related GH family 12 enzymes, and their variants, have provided insight on how specific residues contribute to protein thermal stability and enzyme activity. This knowledge can in the future serve as a structural toolbox, i.e., to design Cel12A enzymes with specific properties and features by introducing subtle changes in structural components of the enzymes. These can then be utilized to develop new industrial products or fine-tune enzymes in already existing applications.

Cyanobacteria are a heterogeneous group of phototrophic microorganisms. Many cyanobacteria have the capacity to fix atmospheric nitrogen. During the process of nitrogen fixation, molecular hydrogen is produced. Three enzymes are directly involved in hydrogen metabolism in cyanobacteria. A nitrogenase, evolving hydrogen during nitrogen-fixation, an uptake hydrogenase, recycling the hydrogen produced by nitrogenase, and a bidirectional hydrogenase that has the capacity to both take up and produce hydrogen. The main objective in this thesis was to examine the transcriptional regulation of both the uptake and the bidirectional hydrogenase in filamentous cyanobacteria. The transcriptional regulation of the uptake hydrogenase was demonstrated to be influenced by external conditions in Nostoc muscorum and Nostoc punctiforme. Nickel, molecular hydrogen, and anaerobic conditions all induced the relative amount of uptake hydrogenase transcript. In addition, a transcript could be detected in nitrogen-fixing, but not in non-nitrogen fixing conditions. The transcriptional regulation of the bidirectional hydrogenase in N. muscorum and Anabaena PCC 7120 was also examined. The relative amount of transcript from the bidirectional hydrogenase in both strains was demonstrated to increase during anaerobic conditions. Moreover, experiments using N. muscorum demonstrated that addition of nickel also increase the amount of transcript. However, no change in the relative amount of transcript from the bidirectional hydrogenase could be observed by additional hydrogen or during a shift from non-nitrogen fixing to nitrogen fixing conditions. The genes responsible for maturation of the hydrogenase were identified, cloned and sequenced in N. punctiforme. The transcription of the genes was examined and all genes were located on a single transcript. Like the uptake hydrogenase, a transcript could be detected under nitrogen-fixing but not under non-nitrogen fixing conditions. Initial studies, using microarrays, were used to analyse and compare the transcription of a large set of Anabaena PCC 7120 genes under non-nitrogen and nitrogen-fixing conditions. Both up- and down-regulated genes could be identified. This thesis advances the knowledge about the transcriptional regulation of the hydrogenases in filamentous cyanobacteria and can be used as a platform for further experiments aiming at a modified hydrogen metabolism.

Heparan sulfate (HS) is a long unbranched polysaccharide composed of alternating glucosamine and hexuronic (glucuronic or iduronic) acid residues. Modification by sulfate groups in various positions generates a highly heterogeneous molecule. HS is synthesized as a proteoglycan by virtually all cells, and play pivotal functions in signaling and developmental patterning, but also in pathogenic events such as tumor metastasis and microbial adhesion. This thesis deals with the properties of enzymes involved in HS chain elongation. Polymerization of the HS chain is believed to be catalyzed by the EXT family of proteins. In humans, the EXT family consists of five members: EXT1, EXT2, EXTL1, EXTL2 and EXTL3; their respective functions in HS biosynthesis are not fully understood. In this study, for the first time, successful in vitro HS polymerization on oligosaccharide acceptor substrates was demonstrated, using recombinant EXT1 and EXT1/EXT2 complex. EXT1 formed longer chains than EXT1/EXT2 and their mechanisms of sugar incorporation were different. Suppression of EXT1 or EXT2 expression by siRNA in a human cell line resulted in reduction of HS chain length. In contrast, cells transfected with EXTL3 siRNA produced longer HS chains. Overexpression of soluble EXT1, alone or co-expressed with EXT2, resulted in increased chain length, whereas overexpression of soluble EXT2 or EXTL3 has no detectable effect on HS chain elongation. Structural analysis of HS from fibroblasts isolated from mice with a hypomorphic mutation in Ext1 showed that they produced significantly shorter HS chains then the wild-type fibroblasts (20 and 70 kDa, respectively). The disaccharide composition of the HS produced by the mutant cells was virtually indistinguishable from that of the wild-type HS, however, the mutant HS chains contained higher proportions of unmodified regions. Mutant cells responded less efficiently than wild-type cells to low concentrations of FGF2, as analyzed by ERK phosphorylation assay.