**3.2. Spectral analysis and dot hybridization**

278 The Complex World of Polysaccharides

**Figure 10.** Hydrogen bonding in the complementary pairs G–glucuronic acid, A–glucuronic acid, and C–N-acetylglucosamine in a DNA–polysaccharide double helix exemplified by a five-unit chain.

Е = -7.04 kcal/mol per complementary pair

The interaction purine–glucuronic acid in the NA–polysaccharide double helix was much the same as in isolation: the lengths of the two hydrogen bonds were 1.79 Å. However, interesting differences were observed for the pyrimidine–N-acetylglucosamine interaction in the double helix and in isolation. In addition to the only hydrogen bond (1.86 Å) formed in isolation (Fig. 10), another hydrogen bond (1.84 Å) was formed with O of the neighboring nucleotide in the double helix. Moreover, there was one more, weaker hydrogen bond (2.59 Å) directed oppositely. Thus, N-acetylglucosamine also forms two hydrogen bonds with nucleotides of an NA strand, which ensures its sufficiently tight and highly selective interaction with nucleotides of a native NA molecule. In addition, computations were performed for incorrect sequences of monosaccharides. As in the case of incorrect sequences of bases in the DNA double helix, low-energy hydrogen bonds were obtained in this variant or the NA–polysaccharide double helix was not formed at all because the bonds were

1. Quantum chemical calculations showed that hydrogen bonds are formed in the complementary pairs purine–glucuronic acid and pyrimidine–N-acetylglucosamine,

2. Geometric parameters of the interacting NA and polysaccharide strands play the major role in selection of particular monosaccharides during putative template synthesis of polysaccharides. The difference observed for the energy of hydrogen bonds is

3. Quantum chemical modeling confirmed that purines are complementary to the carboxyl group and pyrimidines, to the hydroxymethyl group of UDP-

Quantum chemical computation by the PM3 method. (Another view point).

disadvantageous in terms of energy. The conclusions are as follows.

allowing the formation of a NA–polysaccharide double helix.

monosaccharides.

insufficient for selection of monosaccharides with necessary structures.

Table 7 shows the polysaccharide composition determined biochemically. It should be noted that all polysaccharides were end products of modification, that is, mature glycans. Polyuronides, which consist mostly of uronic acid residues, and amylose, which consists only of glucose residues, can be considered as virtually nonmodified glycans. The hydroxymethyl group at C5 of monosaccharide is substituted with hydrogen in glucuronoxylans. HA and CS have the acetoamide group at C2 of hexose residues. In CS, hydrogen of hydroxyl at C6 is substituted with the sulfo group, which provides an additional partial negative charge to the molecule (Table 7.)


Note: UA, uronic acid (on Dishe); NS, neutral sugar (on anthrone method); AAH, aminoacetylhexoses(AAH=1000- UA); OM, other monosaccharides (OM=1000-UA-NS).

**Table 7.** Characterization of the composition of polysaccharides used in experiments

Electrophoretic analysis of polysaccharides used in experiments is illustrated in Fig. 11. It is seen that polysaccharides were sufficiently pure, which was important for our study.

The results shown in Figs. 12A, 12B, and 13A demonstrate that the absorption and CD spectra remained much the same when poly(dC) was combined with oat polyuronides or poly(dA), with amylose. This finding suggests that these polynucleotides and polysaccharides do not form sufficiently tight complexes.

When poly(dA) was combined with oat polyuronides or poly(dC), with amylose, a decrease in amplitude was observed on both the absorption and CD spectra (Figs. 12C, 13B, 13C). The decrease in the peak intensity of the CD spectrum and the simultaneous decrease in the amplitude of peak absorption (a hypochromic effect) can result from an increase in the degree of spiralization and the twist of the polynucleotide (Blagoi Yu.P., et al. 1999), suggesting the formation of a single-stranded polynucleotide–polysaccharide complex. It should be noted that the amplitude of the absorption spectra changed by 9-11% in this case. Since the hyperchromic effect of double-stranded DNA (poly(dA)–poly(dT)) in 0.3 М NaCl was 27-30% (Figs. 14А, 14B), which agreed with published data (Yang J.T., et al. 1969), the change observed in our experiment accounted for 30-35% of the hyperchromic effect characterizing complete base pairing.

**Figure 12.** A. Absorption spectrum of poly(dC) in the absence (curve 1) or presence (curve 2) of polyuronide (PU), r = 2. (r - is hereafter the ratio of polysaccharide units to the molar concentration of nucleotides.) B. CD spectrum of poly(dC) in the absence (curve 1) or presence (curve 2) of PU, r = 2. C. Absorption spectrum of poly(dA) in the absence (curve 1) or presence (curve 2) of PU, r = 1. D. CD spectrum of poly(dA) in the presence (curve 1) or absence (curve 2) of PU, r = 1.

1, oat polyuronides 3, HS 2, garlic glucuronoxylans 4, CS 5, HA

**Figure 11.** Electrophoresis of polysaccharides in cellulose acetate bands in 0.1 N НСl.

1 2 3 4 5

**Figure 12.** A. Absorption spectrum of poly(dC) in the absence (curve 1) or presence (curve 2) of polyuronide (PU), r = 2. (r - is hereafter the ratio of polysaccharide units to the molar concentration of nucleotides.) B. CD spectrum of poly(dC) in the absence (curve 1) or presence (curve 2) of PU, r = 2. C. Absorption spectrum of poly(dA) in the absence (curve 1) or presence (curve 2) of PU, r = 1. D. CD

spectrum of poly(dA) in the presence (curve 1) or absence (curve 2) of PU, r = 1.

**Figure 13.** А. Absorption spectrum of poly(dA) in the absence (curve 1) or presence (curve 2) of amylose (AM), r = 2.B. Absorption spectrum of poly(dC) in the absence (curve 1) or presence (curves 2, 3) of AM; r = 0.5 (1), r = 1 (2), r = 2 (3). C. CD spectrum of poly(dC) in the absence (curve 1) or presence (curve 2) of AM, r = 2.

**Figure 14.** А. Absorption spectra of poly(dA)–(dT)16 (curve 1) and an equimolar mixture of poly(dA) and (dT)16 (curve 2), r =1. B. CD spectra of of poly(dA)–(dT)16 (curve 1) and an equimolar mixture of poly(dA) and (dT)16 (curve 2), r = 1.

The results shown in Figs. 15A and 15B demonstrate that CS does not form a tight complex with DNA, because the DNA absorption spectrum was almost completely restored after annealing in the presence of CS. Another situation was observed with HA (Fig. 15A): a hyperchromic effect of about 10-11% was observed in this case. This finding suggests complexation of HA with DNA regions, because such complexation should prevent the DNA duplex from being restored after annealing in the presence of HA.

**Figure 15.** А. Absorption spectra of calf thymus DNA in the absence (curve 1, before annealing; curve 2, after annealing) or presence (curve 3, after annealing) HA; r = 1. B. Absorption spectra of calf thymus DNA in the presence of CS before (curve 1) and after (curve 2) annealing, r = 1.

Dot hybridization showed that DNAs of all organisms specifically binds with oligodeoxyribonucleotide probes (Fig. 16). Although the method is semiquantitative, it is possible to state that the GC tandem repeat has more complementary sequences in DNA as compared with monopurine (A) or monopyrimidine (C) homopolymers. For instance, poly(dT) and poly(dGC) arrays are rather abundant in human DNA, while poly(dG) occur in a smaller amount. Garlic DNA showed a greater amount of poly(dG) as compared with human DNA. This result agrees with the modern views of the plant genome (Zelenin A.V. 2003).

**Figure 16.** Dot hybridization of polysaccharides and DNA with oligodeoxyribonucleotide probes. The membrane was autoradiographed.


Since we aimed at studying the possibility of specific bonding between NA and polysaccharides, the results obtained with DNA were used exclusively as a control.

282 The Complex World of Polysaccharides

The results shown in Figs. 15A and 15B demonstrate that CS does not form a tight complex with DNA, because the DNA absorption spectrum was almost completely restored after annealing in the presence of CS. Another situation was observed with HA (Fig. 15A): a hyperchromic effect of about 10-11% was observed in this case. This finding suggests complexation of HA with DNA regions, because such complexation should prevent the

**Figure 15.** А. Absorption spectra of calf thymus DNA in the absence (curve 1, before annealing; curve 2, after annealing) or presence (curve 3, after annealing) HA; r = 1. B. Absorption spectra of calf thymus

Dot hybridization showed that DNAs of all organisms specifically binds with oligodeoxyribonucleotide probes (Fig. 16). Although the method is semiquantitative, it is possible to state that the GC tandem repeat has more complementary sequences in DNA as compared with monopurine (A) or monopyrimidine (C) homopolymers. For instance, poly(dT) and poly(dGC) arrays are rather abundant in human DNA, while poly(dG) occur in a smaller amount. Garlic DNA showed a greater amount of poly(dG) as compared with human DNA. This result agrees with the modern views of the plant genome (Zelenin A.V. 2003).

**Figure 16.** Dot hybridization of polysaccharides and DNA with oligodeoxyribonucleotide probes. The

DNA duplex from being restored after annealing in the presence of HA.

DNA in the presence of CS before (curve 1) and after (curve 2) annealing, r = 1.

1. glucuronoxylans 6. garlic DNA 2. polyuronides 7. human DNA 3. CS 8. fish DNA 4. HA 9. calf thymus DNA

membrane was autoradiographed.

5. amylase

**Table 8.** Radioactivity of dots of polysaccharides and DNA after hybridization with purine-pyrimidine DNA probes

The results of autoradiography (Fig. 16, Table 8) demonstrated that the poly(dA) probe specifically hybridized with oat polyuronides to an extent comparable with that of DNA (garlic DNA). The poly(dC) probe specifically hybridized to a high extent with potato amylose. Since oat polyuronides consist almost exclusively of hexuronic acid residues and potato amylose, of hexose residues, we observed significant specific hybridization of pyrimidines (exemplified by C) with polyhexoses and purines (exemplified by A) with polyuronic acids.

The data obtained with glucoronoxylans were interpreted according to the same principle. It is known that glucuronoxylans contains about one carboxyl group per four neutral saccharide residues; i.e., the content of carboxyl groups is three or four times lower than in polyuronides. Hybridization of glucuronoxylans with poly(dA) was indeed about threefold lower than that of polyuronides.

The regular tandem poly(dGC) probe efficiently hybridized with all polysaccharides and especially with GAG. The highest hybridization efficiency was observed with HA. This finding agrees with the tandem structure of HA consists of glucose–glucuronic acid dimers, which interacted with the purine–pyrimidine dimers of the synthetic probe NA.

**Figure 17.** Dot hybridization of polysaccharides and DNAs with calf thymus DNA. The membrane was autoradiographed.

Figure 17 shows the results of dot hybridization of polysaccharides and DNAs of various organisms with fragmented calf thymus DNA. The highest radioactivity was observed in dot 9, i.e., for self-hybridization of calf genomic DNA. In addition, repetitive elements of calf thymus DNA detected homologous sequences in virtually all genomic DNAs examined. Of all polysaccharides, only HA (dot 4) had matching structures in calf thymus DNA and formed specific complexes similar in radioactivity to DNA–DNA complexes.

#### **3.3. Glycan biosynthesis in rat liver**

The profile of elution of rat liver and marker polysaccharides from DEAE cellulose with a NaCl gradient is shown in Fig. 18. As markers, we used homopolymers of uronic acid (oat polyuronides) and glucose (potato amylose).

**Figure 18.** Profiles of elution of rat liver and marker polysaccharides from DEAE cellulose with a NaCl gradient.

As evident from the elution profile, the rat liver contains a complex spectrum of polysaccharides. Analysis with the marker homoglycan amylose, consisting of glucose residues, showed that polymers with a similar structure of neutral sugars are absent from rat liver glycans isolated by our method (such polymers were eluted from the column during washing). Glucose homoglycans were eluted with 0.01 М NaCl. The first peak of liver glycans was detected at 0.02-0.03 М NaCl. With NaCl increasing to 0.15 М, we eluted polysaccharides with a negative charge greater than in polymers of neutral sugars and lower than in the 0.25 М NaCl fraction, which mostly consisted of HA, as commonly known (Zimnitskii A.N., et al. 2004). The fractions 0.7 and 1.5 М NaCl exemplified a classical separation of CS and HS. As expected, highly charged oat polyuronides were eluted at the highest ionic strength, at about 1.2 М NaCl.

284 The Complex World of Polysaccharides

5, potato amylose

gradient.

autoradiographed.

1, garlic glucuronoxylans 6, garlic DNA 2, oat polyuronides 7, fish DNA 3, animal CS 8, phage DNA 4, animal HA 9, calf DNA

**3.3. Glycan biosynthesis in rat liver** 

polyuronides) and glucose (potato amylose).

**Figure 17.** Dot hybridization of polysaccharides and DNAs with calf thymus DNA. The membrane was

1 2 3 4 5 6 7 8 9

Figure 17 shows the results of dot hybridization of polysaccharides and DNAs of various organisms with fragmented calf thymus DNA. The highest radioactivity was observed in dot 9, i.e., for self-hybridization of calf genomic DNA. In addition, repetitive elements of calf thymus DNA detected homologous sequences in virtually all genomic DNAs examined. Of all polysaccharides, only HA (dot 4) had matching structures in calf thymus DNA and

The profile of elution of rat liver and marker polysaccharides from DEAE cellulose with a NaCl gradient is shown in Fig. 18. As markers, we used homopolymers of uronic acid (oat

**Figure 18.** Profiles of elution of rat liver and marker polysaccharides from DEAE cellulose with a NaCl

As evident from the elution profile, the rat liver contains a complex spectrum of polysaccharides. Analysis with the marker homoglycan amylose, consisting of glucose residues, showed that polymers with a similar structure of neutral sugars are absent from

formed specific complexes similar in radioactivity to DNA–DNA complexes.

The results of quantitating uronic acids, hexosamines, and NS in the chromatographic fractions are summarized in Table 9.


**Table 9.** Concentrations (g/ml) of uronic acids, hexosamines, and NS in fractions of liver polysaccharides of intact rats (rats were not subjected to external influence)

In the 0.25, 0.7, and 1.5 М fractions, the ratio uronic acids : hexosamines was about 1:1, which is characteristic of HA, CS, and HS and agrees with published data. A minor amount of NS detected in the 0.7 and 1.5 М fractions could be explained by the presence of a trisaccharide fragment linking GAG with the core protein in proteoglycans.


**Table 10.** Contents (g per g tissue) of NS/uronic acids in the 0.15 M NaCl saccharide fractions of cell nuclei, microsomes, and a homogenate of the liver of intact rats


**Table 11.** Uronic acids per neutral trisaccharide in the 0.15 M NaCl saccharide fractions of cell nuclei, microsomes, and a homogenate of the liver of intact rats

The 0.15 М fraction contained glycans enriched in neutral sugars (Table 9). Three subfractions could be isolated in this fraction: the first peak eluted up to 0.02 M NaCl, the second peak eluted up to 0.06 M NaCl, and a plateau eluted up to 0.15 M NaCl. The primary structure of polysaccharide fragments of this fraction is characterized in Tables 10 and 11. As evident from Table 11, nuclear oligosaccharides had the NS : uronic acid ratio similar to 3:1, suggesting the presence of neutral trisaccharides containing no more than one uronic acid residue. Both microsomal and homogenate fractions harbored neutral oligosaccharides containing more than one uronic acid residue per trisaccharide, the proportion of uronic acids increasing sixfold from the 0.02 to the 0.15 M subfraction. This finding testifies that uronic acid is linked to trisaccharides mostly in cell microsomes. Only one uronic acid residue per trisaccharide is linked in cell nuclei.


Note: Subscripts n and h correspond to the nuclear fraction and the homogenate, respectively.

**Table 12.** Contents (g per g tissue) of DNA and GAG in the nuclear fraction and the homogenate of the liver of intact rats

The DNA content and the GAG composition in the nuclear fraction and the homogenate of the rat liver are shown in Table 12. The total GAG content was 6.12 g per g tissue in the nuclear fraction and 102.81 g per g tissue in the homogenate. Nuclear GAG contained 0.15 М fraction PS to 23.20%, HA to 23.86%, CS to 41.83%, and HS to 11.11%. GAG of the homogenate contained 0.15 М fraction PS to 40.87%, HA to 23.09%, CS to 22.95%, and HS to 13.08%.

Our isolation procedure yielded about 5% of nuclei in the intact form. The remaining nuclei, along with their contents, were in the liver homogenate.

Correlation analysis was performed to study the association between the biochemical parameters and the time after a glucose load. Since the empirical sample was small (n<30), we computed Pearson parametric empirical correlation coefficients and Spearman nonparametric rank correlation coefficients. The results are summarized in Tables 13-15.

As Table 13 shows, a strong positive correlation was observed between time and HS content in the homogenate (r = 0.81) and between the contents of HS in nuclei and the 0.15 M fraction in the homogenate (r = 0.95) and a strong negative correlation between the contents of the 0.15 M fraction and DNA in nuclei (r = -0.71).


Concept of Template Synthesis of Proteoglycans 287

Note: Significant correlation coefficients are in bold.

286 The Complex World of Polysaccharides

the liver of intact rats

residue per trisaccharide is linked in cell nuclei.

Substance Content DNAn 9.26±0.30 0.15 Mn fraction of PS (Fig. 18) 1.42±0.12 HAn 1.46±0.14 CSn 2.56±0.16 HSn 0.68±0.07 DNAh 167.61±6.94 0.15 Мh fraction of PS (Fig. 18) 42.02±4.13 HAh 23.74±3.35 CSh 23.60±3.31 HSh 13.45±1.22 Note: Subscripts n and h correspond to the nuclear fraction and the homogenate, respectively.

**Table 12.** Contents (g per g tissue) of DNA and GAG in the nuclear fraction and the homogenate of

The DNA content and the GAG composition in the nuclear fraction and the homogenate of the rat liver are shown in Table 12. The total GAG content was 6.12 g per g tissue in the nuclear fraction and 102.81 g per g tissue in the homogenate. Nuclear GAG contained 0.15 М fraction PS to 23.20%, HA to 23.86%, CS to 41.83%, and HS to 11.11%. GAG of the homogenate

Our isolation procedure yielded about 5% of nuclei in the intact form. The remaining nuclei,

Correlation analysis was performed to study the association between the biochemical parameters and the time after a glucose load. Since the empirical sample was small (n<30), we computed Pearson parametric empirical correlation coefficients and Spearman nonparametric rank correlation coefficients. The results are summarized in Tables 13-15.

As Table 13 shows, a strong positive correlation was observed between time and HS content in the homogenate (r = 0.81) and between the contents of HS in nuclei and the 0.15 M fraction in the homogenate (r = 0.95) and a strong negative correlation between the contents

contained 0.15 М fraction PS to 40.87%, HA to 23.09%, CS to 22.95%, and HS to 13.08%.

along with their contents, were in the liver homogenate.

of the 0.15 M fraction and DNA in nuclei (r = -0.71).

second peak eluted up to 0.06 M NaCl, and a plateau eluted up to 0.15 M NaCl. The primary structure of polysaccharide fragments of this fraction is characterized in Tables 10 and 11. As evident from Table 11, nuclear oligosaccharides had the NS : uronic acid ratio similar to 3:1, suggesting the presence of neutral trisaccharides containing no more than one uronic acid residue. Both microsomal and homogenate fractions harbored neutral oligosaccharides containing more than one uronic acid residue per trisaccharide, the proportion of uronic acids increasing sixfold from the 0.02 to the 0.15 M subfraction. This finding testifies that uronic acid is linked to trisaccharides mostly in cell microsomes. Only one uronic acid

**Table 13.** Pearson empirical correlation coefficients (r) between the time after a glucose load and the DNA and GAG contents in the rat liver


Note: Significant correlation coefficients are in bold.

**Table 14.** Spearman rank correlation coefficients (R) between the time after a glucose load and the DNA and GAG contents in the rat liver

Table 14 shows three strong correlations: a negative one between the contents of the 0.15 M fraction and DNA in nuclei and positive correlations between the HS content in nuclei and the CS content in the homogenate and between the contents of the 0.15 M fraction and CS in the homogenate (R = 0.74).

To identify latent variables uniting the parameters under study, factor analysis was performed using the Varimax procedure (Table 15).



Note: Significant correlations between the parameter and the factor are in bold.

**Table 15.** Factor analysis (Varimax) of the time after a glucose load and the DNA and GAG contents in the rat liver

As evident from Table 15, Factor 1 included CSn, HSn, the 0.15 Mh fraction, and CSh. Factor 2 included the 0.15 Mn fraction and HAn. Factor 3 included DNAh and HAh.

The most important result of correlation analysis is that strong linear correlations were observed between the time after a glucose load and the content of HS in the liver homogenate and between the contents of the 0.15 M fraction and DNA in the nucleus. In the first case, we record an increase in the time of synthesis of GAG modified to the greatest extent (at least two sulfo groups per disaccharide fragment). This result was conformed by factor analysis, since the time and the HS content in the homogenate are components of one factor. In the second case, we possibly establish a functional relationship between synthesis of nuclear oligosaccharides contained in the 0.15 M fraction and the genetic apparatus of the cell. The two biopolymers involved belong to different classes, but their biosynthesis utilizes the same initial substrate, glucose. The strong negative correlation suggests competition for the initial substrate between the two biosynthetic pathways. Since the nuclear DNA content was constant under the conditions of our experiments, the negative correlation suggests that synthesis of saccharides belonging to the 0.15 M fraction increases when free glucose enters cells in considerable amounts. This assumption is supported by the results shown in Fig. 19.

The composition of the 0.15 M fraction and GAG in nuclei, microsomes, and the homogenate is shown in Fig. 20. It is seen that nuclear GAG are enriched in СS (about 42%), which agrees with the accepted views (Silbert J.E. Sugumaran G ., 1995). Homogenate GAG consist mostly of HA and CS (about 23%). In microsomes, all GAG types occur in nearly equal proportions (about 12%). The results obtained for the 0.15 M fraction, which consists of three main subfractions (Table 11), are shown in Fig. 21. The fraction occurred at a high content (about 60%) in microsomes, while its content in the nucleus was no more than 23%. The homogenate was intermediate between nuclei and microsomes in the content of this fraction.

288 The Complex World of Polysaccharides

Variance portion accounted for by the

Note: Significant correlations between the parameter and the factor are in bold.

supported by the results shown in Fig. 19.

factor

the rat liver

Parameter Factor 1 Factor 2 Factor 3 t (min) -0.490 0.494 -0.473 DNAn 0.656 0.656 -0.189 0.15 Mn -0.064 **-0.851** -0.009 HAn 0.054 **-0.768** 0.019 CSn **0.804** -0.064 -0.063 HSn **0.913** 0.061 0.361 DNAh 0.386 -0.106 **0.728**  0.15 Мh **0.863** 0.159 0.279 HAh -0.112 0.001 **0.866**  CSh **0.776** -0.500 -0.041 HSh -0.146 0.692 -0.573 Proper factor value 3.687 2.762 2.083

**Table 15.** Factor analysis (Varimax) of the time after a glucose load and the DNA and GAG contents in

As evident from Table 15, Factor 1 included CSn, HSn, the 0.15 Mh fraction, and CSh. Factor

The most important result of correlation analysis is that strong linear correlations were observed between the time after a glucose load and the content of HS in the liver homogenate and between the contents of the 0.15 M fraction and DNA in the nucleus. In the first case, we record an increase in the time of synthesis of GAG modified to the greatest extent (at least two sulfo groups per disaccharide fragment). This result was conformed by factor analysis, since the time and the HS content in the homogenate are components of one factor. In the second case, we possibly establish a functional relationship between synthesis of nuclear oligosaccharides contained in the 0.15 M fraction and the genetic apparatus of the cell. The two biopolymers involved belong to different classes, but their biosynthesis utilizes the same initial substrate, glucose. The strong negative correlation suggests competition for the initial substrate between the two biosynthetic pathways. Since the nuclear DNA content was constant under the conditions of our experiments, the negative correlation suggests that synthesis of saccharides belonging to the 0.15 M fraction increases when free glucose enters cells in considerable amounts. This assumption is

The composition of the 0.15 M fraction and GAG in nuclei, microsomes, and the homogenate is shown in Fig. 20. It is seen that nuclear GAG are enriched in СS (about 42%), which agrees with the accepted views (Silbert J.E. Sugumaran G ., 1995). Homogenate GAG consist mostly of HA and CS (about 23%). In microsomes, all GAG types occur in nearly equal proportions

2 included the 0.15 Mn fraction and HAn. Factor 3 included DNAh and HAh.

0.335 0.251 0.189

**Figure 19.** Accumulation of polysaccharides (PS) of the 0.15 M fraction in the nucleus. The content of the fraction was normalized with respect to the DNA content.

**Figure 20.** Relative contents (%) of polysaccharides in nuclei, microsomes, and the homogenate of the rat liver.

**Figure 21.** Relative contents (%) of polysaccharides differing in the portion of neutral sugars in the 0.15 M fractions of nuclei, microsomes, and the homogenate of the rat liver.

To study biosynthesis of polysaccharide fragments, we analyzed the time course of incorporating radiolabeled glucose in polysaccharide chains (Fig. 22).

**Figure 22.** Growth to the maximal specific radioactivity of polysaccharides as dependent on the time of label presence in rats. The maximal specific radioactivity was taken as 100% for each fraction.

The highest radioactivity of the 0.15 M fraction was detected 60-75 min after a glucose load, while mature GAG (fraction 0.25-1.5 М) showed the highest radioactivity after 120 min. Thus, initiation of glycan synthesis is probably associated with the 0.15 М fraction, enriched in neutral oligosaccharides, which is consistent with common views of proteoglycan biosynthesis.

To identify the subcellular structures where generation of a polysaccharide chain is initiated, we studied the dynamics of label incorporation in the nuclear and microsomal fractions of the rat liver. The results of this experiment are shown in Fig. 23.

**Figure 23.** Dynamics of accumulation of radiolableled glucose in the nuclear and microsomal fractions of the rat liver. As 100%, we used the radioactivity averaged over all fractions and all time points.

As Fig. 23 demonstrates, the label became detectable in cell structures 30 min after a glucose load. The label accumulated to the highest content in both fractions within 1 h of the experiment. It should be noted that, in the first hour, the accumulation rate was higher in the nuclear fraction (radioactivity 53% compared with 44% in the microsomal fraction). After 120 min of the experiment, the microsomal fraction accumulated a greater amount of the label as compared with nuclei (29 and 12%, respectively). After 3 h, the label was again accumulated more intensely in nuclear structures (33% compared with 26% in microsomes).
