Studies on the Estimation and Decomposition of Amino Sugars in Soil

[ 152 ]

STUDIES ON THE ESTIMATION AND DECOMPOSITION OF AMINO SUGARS IN SOIL

BY J. M. BREMNER AND K. SHAW, Rothamsted Experimental Station, Harpenden, Herts

(With Two Text-figures)

Very little is known about the chemical nature of the and sample 3 from a small experimental plot re- nitrogenous organic complexes in soil. Recent receiving annual additions of nitrogen, phosphorus views of the literature on the subject (Kojima, 1947; and potassium fertilizers but no organic manure. Bremner, 1951) show that although the protein Sample 4 was taken from a typical neutral fen soil nature of 30-40 % of soil nitrogen has been esta- in the Cambridgeshire fen district. Sample 5 was blished, the chemical nature of most of the non- from a peat overlying a lacustrine deposit of calcium protein nitrogen is still obscure. Evidence that some carbonate, sedge (Carex flacca, O. paniculata and C. of the non-protein nitrogen is in the form of amino spicata) being the dominant vegetation. Sample 6 Sugars has been provided by the finding that soil was taken from the subsurface (9-15 in.) layer of hydrolysates give positive colour tests for 2-amino a profile developed under vegetation consisting sugars and by the detection of glucosamine on paper chiefly of Molinia and Nardus spp. The pH values of chromatograms of soil hydrolysates (Bremner, 1949, the samples were determined with the glass elec- 1950). However, no reliable information regarding trode, nitrogen contents by a micro-Kjeldahl pro- the proportion of soil nitrogen that occurs in the cedure and CaCO3 contents by the Schollenberger form of amino sugars is yet available. Results (1930) technique (Table 1). obtained by Bremner (1949), using modifications of Hydrolysis of the soils. The method adopted for the Elson & Morgan (1933) and Zuckerkandl & hydrolysis was essentially that used by Smithies Messiner-Klebermass (1931) methods of estimating (1952) for the hydrolysis of hexosamine complexes amino sugars, indicated that the proportion was in fungal mycelium. Each soil sample was allowed small, but no reliance was placed on these analyses to stand with 12N-HC1 (10 ml./g. of air-dried soil) since considerable difficulties were experienced with at room temperature for 2-3 days and then diluted the methods and their applicability to soil was with an equal volume of water and heated in a questionable. Since then these colorimetrie methods sealed tube or under reflux in boiling water for 6 hr. of determining amino sugars have been investigated The mixture was then cooled, filtered through a and improved (Blix, 1948; Ogston & Stanier, 1950; Buchner funnel with suction and the residue washed Immers & Vasseur, 1950, 1952; Johnston, Ogston & thoroughly with distilled water. Aliquots of the Stanier, 1951; Schloss, 1951; Aminoff, Morgan & combined filtrate and washings were taken for total - Watkins, 1952; Smithies, 1953) and a new alkaline N determinations by a micro-Kjeldahl procedure decomposition method of estimation has been de- and for amino sugar-N estimations and paper veloped (Tracey, 1951, 1952). The first part of this chromatographic examination by methods de- paper deals with the application of these methods of scribed below. Tests carried out with soil 6 showed estimating amino sugars to soil. The second section that the amount of amino sugar-N liberated, as gives the results obtained in a study of the decom- determined by the methods described below, was position of amino sugars when added to soil. not increased significantly by prolonging the hydrolysis procedure beyond 6 hr.

A. ESTIMATION OF AMINO Amino sugar estimations SUGARS IN SOIL Alkaline decomposition methods. From Morgan's (1936) observation that practically quantitative re- Materials and methods covery of glucosamine- or galactosamine-N as NH3 Soils. The materials used were as diverse as can be obtained by heating for a few minutes with possible and consisted of three mineral (nos. 1-3) and normal alkali at 100° C, Tracey (1951) has recently three organic soils (nos. 4-6). Sample 1 was from developed an alkaline decomposition method of a continuous wheat plot (Broadbalk 2B) receiving estimating amino sugars which involves distillation farmyard manure annually, sample 2 from an old with saturated NajPO! solution plus sodium borate garden soil that had been cultivated with large (pH 11-1) in a Markham (1942) still and estimation additions of organic manure for almost 100 years, of the NH3 liberated by Nesslerization. In a pre-

liminary investigation of this method it was found a Conway (1947) unit, and the NH3 liberated over- that quantitative recovery of glucosamine-N as NH8 night at room temperature was absorbed in could not be obtained by distilling with N-NaOH or 0-02N-H2SO4 in the inner chamber and determined XajPO4—borate buffer in the Markham apparatus. by Nesslerization. Under the conditions of this It was found, however, that practically quantitative microdiffusion method no decomposition of glucos- recovery of glucosamine-N as NH3 was obtained by amine could be detected, and of the compounds a borate buffer (pH 8-8) distillation method pre- tested for interference in the buffer distillation viously used by one of us (K.S.) for the determina- method only asparagine and glutamine gave tion of ammonia-N in soil extracts. In this method measurable quantities of ammonia, their recoveries an al iquot of the so i 1 extract, containing 0 • 1-1 • 0 mg.being 0-8 and 2-0 % respectively. of ammonia-N, is pipetted into a 500 ml. Kjeldahl Tests showed that no NH3 was liberated when flask, and after adjustment of the pH to 6-3 by the glucosamine was distilled with MgO at 40° in vacuo cautious addition of 2N-NaOH using methyl red and that the results obtained when ammonia-N in as indicator, 50 ml. of a solution containing 8 g. of the hydrolysates was determined by this method boric acid plus 40 g. of sodium borate per litre are were in close agreement with those obtained by the added, and the total volume made up to 300ml. with MgO cold microdiffusion technique. It was found, water. The flask is then connected via a splash trap however, that glucosamine was extensively decom- to a vertical Liebig condenser and the NH3 liberated posed when distilled with MgO at 100° C, recovery by boiling (30 min.) is collected in 2 % (w/v) boric of glucosamine-N as NH3 after 45 min. distillation acid containing 10 ml./l. of Conway & O'Malley's being 86%. (1942) indicator and determined by titration with At a later stage in this work Mr M. V. Tracey kindly O-OIN-HJSO,,. Under the conditions of this buffer - let us have, prior to publication (Traeey, 1952), the

Table 1. Description of soils studied Percentage on moisture-free basis

No. Soil Site pH N CaCO, 1 Clay loam Rothamsted 7-7 0-23 2-39 2 Old garden Rothamsted 6-9 0-43 1-29 3 Medium sandy loam Wolverhampton 6-3 0-17 0 4 Fen Littleport 7-1 1-50 304 5 Low-moor peat Anglesey 6-3 2-82 0 6 Mountain peat Capel Curig 40 2-07 0

distillation method no NH3 or acid-titratable base MS. of the paper giving full details of his Na3PO4- was evolved from the following compounds: serine, borate buffer method of estimating amino sugars threonine, citrulline, arginine, lysine, thymine, using the Markham microdistillation apparatus. uracil, guanine, adenine, xanthine, allantoin, allo- When these details were followed satisfactory re- xan, creatinine, creatine, choline, nicotinamide, coveries of glucosamine-N as NH3 were obtained, hydantoin, ethanolamine, betaine and chitin. NH3 and this method was also used to estimate amino was obtained from urea, asparagine and glutamine, sugars in the soil hydrolysates, samples freed from recoveries being 1-0, 2-6 and 17-8% respectively, HCl by repeated distillation at room temperature but interference due to any free or combined in vacuo over KOH being used. The method used by asparagine and glutamine present in soil would be Tracey to determine preformed NH3, i.e. distillation eliminated, and any urea present would be largely in a Conway microdiffusion unit with half-saturated destroyed, by the acid hydrolysis procedure. Tests K2CO3 for 4 hr. at 5° C, was not adopted. The MgO showed that amino acid-sugar mixtures interfered microdiffusion method described above was pre- in this method as in the alkaline decomposition ferred because it need not be carried out below room method of Tracey (1951) and the colorimetric method temperature to prevent decomposition of glucos- of Elson & Morgan (1933), but such mixtures were amine. not detectable in the soil hydrolysates by the tech- Colorimetric methods. Colorimetric estimations niques described by Immers & Vasseur (1950,1952). were made on samples of hydrolysate freed from (See Results.) HCl by repeated evaporation in vacuo over KOH at The above method was used for the estimation of room temperature by the method of Elson & Morgan amino sugar-N in the soil hydrolysates. Inter- (1933) modified according to Blix (1948) and ference due to preformed NH3 in the hydrolysates Immers & Vasseur (1950) and by the method of was allowed for by separate estimations using a Zuckerkandl & Messiner-Klebermass (1931) modi- microdiffusion technique in which MgO was added fied according to Morgan & Elson (1934) and to an aliquot of hydrolysate in the outer chamber of Smithies (1953).

Table 2. Recovery of amino sugar-'N after hydrolysis of chitin and glucosamine in the presence and absence of soil for 6 hr. at 100° C. Bosulta in mg. of N, except in last column.) Becovery of added amino sugar-N <%) Total- Soluble- Distill- Ammonia- Amino Soil Soil Material N N able-N"» Nf augar-NJ absent present Chitin 1. Chitin (0-2 g.) 12-12 1210 10-58 0-84 9-74 80-4 — 2. SoU 3 (2-0 g.) 3-40 2-60 0-78 0-63 015 — 3. Chitin (0-2 g.) + soil 3 (20 g.) 15-52 14-70 1119 1-26 9-93 4. Chitin, from 3-2 12-12 11-90 10-41 0-63 9-78 — 80-8 5. Soil 4 (1-5 g.) 20-52 12-53 3-55 2-43 112 . . 6. Chitin (0-2 g.) + soil 4 (1-5 g.) 32-64 24-60 14-24 3-40 10-84 7. Chitin, from 6-5 1212 1207 10-69 0-97 9-72 — 80-2 8. Soil 6 (1-5 g.) 25-62 16-71 5-42 305 2-37 — — 9. Chitin (0-2 g.) + soil 6 (1-5 g.) 37-74 28-53 1612 3-84 12-28 10. Chitin, from 9-8 1212 11-82 10-70 0-79 9-91 — 81-8 Glueosamine 11. Glucosamine (0-2 g.) 12-98 12-63 11-80 0-52 11-28 86-9 — 12. Glucosamine (0-2 g.) + soil 3 (20 g.) 16-38 15-01 12-70 1-24 11-46 — 13. Glucosamine, from 12-2 12-98 12-41 11-92 0-61 11-31 — 871 14. Glucosamine (0-2 g.) + soil 6(1-5 g.) 38-60 29-33 17-18 3-72 13-46 — 15. Glucosamine, from 14-8 12-98 12-62 11-76 0-67 1109 — 85-5

* Distillable-N represents (amino sugar + ammonia)-N liberated as NH3 by distillation with borate buffer (pH 8-8). •f Estimated by distillation with MgO at room temperature in Conway microdiffusion units, j Distillable-N minus ammonia-N.

solvent an attempt was made to establish the this basis he estimated the amount of amino sugar-N identity of the amino sugar material present in the in a fungal mycelium by applying a correction factor soil hydrolysates by running chromatograms in sol- of 1-25 to the amount of amino sugar-N found after vents reported to effect a separation of the two hydrolysis. The same correction factor has been hexosamines, namely s-collidine and butanol- applied to the results in Table 3 which shows the ammonia (Aminoff & Morgan, 1948) and n-butyl amounts of amino sugar-N present in the soil acetate-acetic acid (Yosizawa, 1950). However, the hydrolysates as estimated by the alkaline decom- separations achieved were too unsatisfactory to position and eolorimetric techniques. permit any conclusion regarding the identity of the The Elson & Morgan (1933) method, which de- amino sugar material. pends upon the formation of a red colour when 2-amino sugars are heated with an alkaline (NajCOj) Results solution of acetylacetone and then treated with The decomposition of chitin and glucosamine Ehrlich's reagent (p-dimethylaminobenzaldehyde) under the conditions adopted for hydrolysis of the in acid solution, is known to be subject to inter- soils and the effect of soil upon their decomposition ference by acetylglucosamine and certain pyrrole during hydrolysis were studied by determining the and indole derivatives (Elson & Morgan, 1933), urea

Downloaded from https://www.cambridge.org/core. BBSRC, on 10 Sep 2019 at 09:35:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0021859600046232 J. M. BREMNEB AND K. SHAW 155 and glyceraldehyde (Lutwak-Mann, 1941), protein Application of the Elson & Morgan method to the (Johns & Marrack, 1952) and certain amino acid- hydrolysates of the three mineral soils (nos. 1—3) was sugar mixtures (Sideris, Young & Krauss, 1938; rendered difficult by the development of an orange- Bendich & Chargaff, 1946; Vasseur & Immers, 1949; red colour on addition of the acetylacetone reagent Horowitz, Ikawa & Fling, 1950; Gottschalk & and by cloudiness in the solutions after addition of Partridge, 1950a, b; Immers & Vasseur, 1950,1952). Ehrlich's reagent. This was presumably due, at Interference due to pyrrole or indole derivatives is least in part, to the presence of considerable amounts readily detected since these condense with Ehrlich's of ferric iron in these hydrolysates, the ferric ion reagent in acid solution to yield coloured solutions forming a coloured (yellow to orange-red) co- without previous heating with the acetylacetone ordination complex with acetylacetone (Mellan, reagent. Tests showed that such substances were not 1941). The addition of sodium fluoride was found to present in the soil hydrolysates. Of the other inter- eliminate this cloudiness and to reduce the intensity fering substances listed only amino acid-sugar of the interfering colour. Amino sugar-N in these mixtures had to be considered as a possible source of hydrolysates was therefore estimated by a modified serious error in the estimation of amino sugars in soil method in which sodium fluoride (1 ml. of 1 %, w/v) hydrolysates by the Elson & Morgan method. Such was added to the test solutions before addition of the

Table 3. Amounts of distillable-TX', ammonia-'N and amino sugar-N present in the soil hydrolysates and the estimated amounts of amino sugar-~N present in the soils (Results expressed as percentages of the total soil-N.) Amino sugar-N in hydrolysates Distillation Colorimetric D is tillable-N* Ammonia- methods methodsj Amino Soil Nf sugar-N no. A B C A-C B-C D E Mean in soils§ 1 19-3 211 14-6 4-7 6-5 5-7 — 5-6 7-0 o 17-6 17-8 13-7 3-9 41 3-8 — 3-9 4-9 3 21-2 22-9 161 51 6-8 6-7 5-4 60 7-5 4 17-5 17-8 11-8 5-7 60 4-9 60 5-7 7-1 5 18-6 191 11-6 7-0 7-5 5-1 41 5-9 7-4 6 21-2 21-5 11-8 9-4 9-7 7-6 60 8-2 10-2

* Distillable-N represents (amino sugar + ammonia)-N liberated as NH3 by distillation with A, pH 8-8 borate buffer; B, pH 11-1 Na,P04-borate buffer (Tracey, 1952). f Estimated by distillation with MgO at room temperature in Conway microdiffusion units. % D, method of Elson & Morgan (1933) modified according to Blix (1948) and Immers & Vasseur (1950); E, method of Zuckerkandl & Messiner-Klebermass (1931) modified according to Morgan & Elson (1934) and Smithies (1953). § Calculated from the mean values for the amino sugar-N contents of the hydrolysates by applying a correction factor (1-25) to allow for loss of amino sugar-N during hydrolysis.

mixtures also interfere with the estimation of amino acetylacetone reagent and the interfering colour not sugars by alkaline decomposition (Tracey, 1952) but eliminated by fluoride was allowed for by running not with their estimation by the method of Zucker- controls containing all the reagents except p-d\- kandl & Messiner-Klebermass (Smithies, 1953). methylaminobenzaldehyde. Although the results Immers & Vasseur (1950,1952) have shown that the obtained by this method were consistent and tests colour reaction given by amino acid-sugar mixtures showed that the addition of fluoride did not inter- may be differentiated from that due to amino sugars fere with the estimation of glucosamine by the by a relatively simple test based on the fact that Elson & Morgan method, it is clear that estimates glucosamine does not give a colour with Ehrlich's obtained in this way are less certain than those reagent after being heated with NajCOj solution in obtained by the other methods employed. the absence of acetylacetone whereas the amino No difficulty of this nature was experienced with acid-sugar mixtures known to interfere give a red the Zuckerkandl & Messiner-Klebermass method, colour. Since no colour reactions were obtained which involves acetylation of the amino sugars when this test was applied to the soil hydrolysates present and subsequent estimation of the colour pro- suoh mixtures were either completely absent or duced when the N-acetyl amino sugar is heated in present in amounts too small to be detected. This alkaline solution and treated with Ehrlich's reagent. result was confirmed by other tests described by It was found, however, that this method could not Immers & Vasseur (1950) for the detection of inter- be applied to the hydrolysates of soils 1 and 2, since ference due to amino acid-sugar mixtures. substances in these hydrolysates interfered with the

Downloaded from https://www.cambridge.org/core. BBSRC, on 10 Sep 2019 at 09:35:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0021859600046232 156 Estimation and decomposition of amino sugars in soil acetylation of the amino sugars present. Recovery (1941). Ammonia-N was extracted with N-K2iSO4 by this method of glucosamine added to the other containing enough H2SO4 to bring the pH of the hydrolysates (nos. 3-6) was practically quantitative. extract to 1-0 and determined in the extract by the The possibility that soil hydrolysates yield vola- microdiffusion method described in the previous tile amines that are distilled with, and estimated as, section. ammonia when (amino sugar + ammonia)-N is estimated by the pH 8-8 borate buffer distillation Results method was investigated by methods already The amounts of ammonia- and nitrate-N found described (Bremner, 1949). No interference due to after the various periods of incubation are given in liberation of such amines could be detected. Table 4. Only trace amounts (< 1 p.p.m.) of nitrite-N could be detected. B. DECOMPOSITION OF AMINO SUGARS The increases in nitrate-N and (ammonia+ IN SOIL nitrate)-N over the amounts found in the control experiments are expressed graphically in Figs. 1 The decomposition of amino sugars in soil was and 2 respectively. Since nitrogen was added to the studied by comparing the rates of decomposition of soil at the rate of 100 p.p.m. these increases repre- chitin, glucosamine, casein and yeast nucleic acid sent the percentages of added nitrogen recovered as when incubated with soil under conditions found to nitrate-N and (ammonia+ nitrate)-N. produce rapid nitrification of ammonium sulphate. Decomposition was followed by determining the amounts of ammonia-, nitrite- and nitrate-N present DISCUSSION after various periods of incubation. It is obvious that the amount of amino sugar-N Materials and methods present in a complex material such as soil cannot be The soil used, a slightly acid (pH 6-1) medium reliably estimated by any one method of amino sandy light loam containing 0-103 % nitrogen, was sugar analysis. The possibilities of interference aro selected because it had a satisfactory nitrifying so great that it is only by the application of several capacity and did not fix added ammonia in a non- different methods of analysis that a reliable estimate exchangeable form. The air-dried soil, which con- is likely to be obtained. The results obtained by the tained 13 p.p.m. of (ammonia + nitrate)-N, was methods of analysis used in this investigation, which ground to pass a 2 mm. sieve and 60 g. samples of are based on three different principles, are not the sieved material were weighed into 250 ml. greatly different (Table 3). Estimates obtained by beakers. Calcium carbonate (0-1 g.), nitrogen the colorimetric procedures were generally lower (100 p.p.m. of oven-dried soil) in the form of chitin, than those obtained by the buffer distillation tech- casein, etc., and water (6 ml.) were then added, the niques, the alkaline decomposition method of contents of the beakers being mixed thoroughly Tracey (1952) giving the highest results throughout. after each addition. In the case of ammonium sul- Since it is impossible to decide which of the methods phate the requisite amount was dissolved in the employed gives the most accurate estimate of amino 6 ml. of water added. The amounts of calcium car- sugar-N, the amounts of amino sugar-N in the soils bonate and water used were those found in pre- have been calculated from the mean estimates of liminary work to produce optimum, conditions for amino sugar-N in the hydrolysates by applying a nitrification at 25° C, the pH of the soil after correction factor of 1*25 to allow for loss of amino addition of the CaCO3 being 7-6. After the additions sugar-N during hydrolysis. The use of this cor- the samples were transferred to 500 ml. Erlenmeyer rection factor, which is based on the recovery of flasks which were then stoppered and incubated at amino sugar-N after hydrolysis of chitin (Table 2), 25° C. for periods of 10, 20, 35, 50 and 70 days. involves the assumption that the amino sugar com- Control experiments, in which no nitrogen was added plexes of soil resemble chitin in their decomposition to the soil, were run simultaneously. All flaskswer e on acid hydrolysis. However, some correction for aerated at 4-day intervals during incubation. At loss of amino sugar-N on hydrolysis is clearly the end of an incubation period the appropriate required, and it is unlikely that a correction based flasks were removed from the incubator and the on this assumption introduces any serious error. entire contents of each flask used for moisture, Confidence in these estimates is strengthened by the ammonia-, nitrite- and nitrate-N determinations. fact that the recovery of amino sugar-N after hydro- Nitrate-N was extracted by the calcium sulphate lysis of chitin or glucosamine was not affected by method of Roller & McKaig (1939) and determined the presence of soil during hydrolysis (Table 2) and colorimetrically by means of phenol disulphonic that substances known to interfere with the methods acid. Nitrite-N was determined in the same extract of analysis employed could not be detected in the by the colorimetric method described by Shinn soil hydrolysates.

Downloaded from https://www.cambridge.org/core. BBSRC, on 10 Sep 2019 at 09:35:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0021859600046232 J. M. BREMNER AND K. SHAW 157 The finding that 5-10 % of the total nitrogen in amino sugars, which are known to be readily decom- the soils examined was in the form of amino sugars posed in alkaline solution with formation of am- is of considerable interest in connexion with results monia, raised the possibility that some of the obtained in studies on the nitrogen distribution of ammonia reported to be present in soil hydrolysates soil hydrolysates. It has been customary in such is formed during estimation by alkaline decom- studies to estimate ammonia-N by distilling with position of amino sugars. This possibility was

Table 4. Amounts of ammonia-N and nitrate-'N, expressed as p.p.m. of oven-dried soil, found after various periods of incubation Time in days ... 10 20 35 50 70

Nitrogenous material NH,-N N03-N NH3-N N03-N NH3-N N03-N NH3-N NO3-N NH3-N N03-N added* —t 1 13 2 18 2 23 3 21 2 24 ChitinJ 2 22 2 46 2 65 4 64 2 72 (51ucosamine§ 34 31 2 70 2 76 4 74 2 88 Casein|| 18 59 3 87 2 92 4 94 2 97 Nucleic acid^f 32 54 3 93 2 97 4 97 0 106 Ammonium sulphate 25 84 2 112 2 117 3 114 2 115 * N added at rate of 100 p.p.m. of oven-dried soil. t Control. % Sample of locust chitin (6-7 % N) kindly supplied by Mr M. V. Tracey. § Added in the form of its hydrochloride (Roche Products Ltd.). || Fat- and vitamin-free product obtained from British Drug Houses Ltd. If Preparation containing 15-8 % N obtained from Light and Co.

100

Z~70 •o o 160

z i 40

2 30

10 20 30 40 50 60 70 Days

Fig. 1. Percentagge recoveries of added nitrogeg n as Figg. 2. Percentage recoveries of added nitrogen as nitrate-N after various periods of incubation. (ammonia+ nitrate)-N after various periods of in- # 0, chitin; •-•-#, glucosainine; O O, cubation. • •, chitin; • •, glucosamine; nucleic acid; O - - - O, casein; A A, ammonium O O» nucleic acid; O - - - - O. casein; A A, sulphate. ammonium sulphate. MgO (or CaO) at low temperature under reduced eliminated by the tests which showed that no pressure, and it has been concluded from the results ammonia is formed from glucosamine under the obtained in this way that a large proportion—as conditions of the method generally used to estimate much as 25 %—of the organic nitrogen of soil is ammonia-N in soil hydrolysates, i.e. by distillation liberated as ammonia by acid hydrolysis. The origin with MgO in vacua at 40° C. However, the results of this ammonia has been the subject of much presented in Table 2, which confirm previous speculation. The fact that soil hydrolysates contain evidence that glucosamine is unstable towards acid

Downloaded from https://www.cambridge.org/core. BBSRC, on 10 Sep 2019 at 09:35:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0021859600046232 158 Estimation and decomposition of amino sugars in soil hydrolysis (Ogston & Stanier, 1950; Folkes, Grant & lation of ammonia was observed during the initial Jones, 1950; Smithies, 1952), show that some of the rapid decomposition of glucosamine, casein and ammonia found in soil hydrolysates is formed by yeast nucleic acid, but not in subsequent stages of decomposition of amino sugars during hydrolysis. their decomposition or at any stage in the decom- The finding that although glucosamine is not de- position of chitin. It is of interest to note that the composed when distilled with MgO at 40° C. in vacuo recoveries as nitrate of added organic nitrogen after it is extensively decomposed when distilled with 70 days are inversely related to the C/N ratios of the MgO at 100° C. is of interest in connexion with some nitrogenous substances added. The same relation- results obtained by Kojima (1947) when studying ship between the rates of nitrification of various the nitrogen distribution of an acid hydrolysate of compounds, particularly amino-acids, and their a muck soil. Kojima found that the 'basic-N' C/N ratios has been noted by other workers (Batham, fraction of this hydrolysate, i.e. the fraction pre- 1927; Quastel & Scholefield, 1949; Owen, Winsor & cipitated by phosphotungstic acid after removal of Long, 1950). NH3 by distillation with MgO at 45-50° C. under reduced pressure, yielded a considerable amount of SUMMARY volatile base-N when distilled with MgO at 100° C. It seems very likely that this volatile base-N con- 1. The amounts of amino sugar-N present in acid sisted largely of NH3 formed by alkaline decom- hydrolysates of six soils with nitrogen contents position of amino sugars present in the basic ranging from 0-17 to 2-82 % have been estimated by fraction. If it is assumed that all of the volatile colorimetric and alkaline decomposition methods. base-N liberated on distillation of the basic fraction 2. Recovery of amino sugar-N after hydrolysis of with MgO at 100° C. was derived from amino sugars, chitin or glucosamine was found to be unaffected by then it can be calculated from Kojima's results that the presence of soil during hydrolysis. at least 9-3 % of the total-N in the soil examined was 3. Substances known to interfere with the in the form of amino sugars. methods of amino sugar analysis employed were not Waksman (1938), in his monograph on soil organic detectable in the soil hydrolysates. matter, states that' the presence of chitin in forest 4. From the amounts of amino sugar-N liberated soils was first demonstrated by P. E. Muller (1887)' by acid hydrolysis it is deduced that 5-10 % of the and that 'H. von Post (reported by Ramaan, 1888) total-nitrogen of the soils examined was in the form found that as much as 15 to 20 per cent of the dry of amino sugars. weight of peat may consist of chitin'. These state- 5. The decomposition of amino sugars in soil has ments are very misleading. Reference to the original been studied by comparing the rates of decom- literature shows that the material referred to as position of chitin, glucosamine, casein and yeast ' chitin' was merely the skeletal remains of insects. nucleic acid when incubated with soil under con- The results obtained in the decomposition experi- ditions found to produce rapid nitrification of ments (Table 4; Figs. 1,2) show that although chitin ammonium sulphate. and glucosamine undergo considerable decom- 6. Glucosamine and chitin are readily decomposition when added to soil their rates of decomposed by soil micro-organisms but not so rapidly as position are not so rapid as those of casein and yeast casein or yeast nucleic acid. nucleic acid. Glucosamine, like casein and yeast nucleic acid, was decomposed rapidly in the first The authors are indebted to Mr M. V. Tracey and 10 days of incubation and then very slowly over the Dr W. R. Smithies for help and advice during the following 60 days. In comparison the decomposition course of this work and to Mr R. G. Warren for of chitin was slow, proceeding at a steady rate during helpful suggestions in the preparation of the the first 35 days and thereafter very slowly. Accumu- manuscript.

REFERENCES

AMINOFF, D. & MORGAN, W. T. J. (1948). Nature, Lond., BREMNER, J. M. (1951). J. Soil Sci. 2, 67. 162, 579. CONSDEN, R., GORDON, A. H. & MARTTN, A. J. P. (1947). AMTNOFF, D., MORGAN, W. T. J. & WATKINS, W. M. Biochem. J. 41, 590. (1952). Biochem. J. 51, 379. CONWAY, E. J. (1947). Microdiffusion Analysis and BATHAM, H. N. (1927). Soil Sci. 24, 187. Volumetric Error, 2nd ed. London: Crosby Look- BENDICH, A. & CHARGAFF, E. (1946). J. Biol. Ohem. 166, wood. 283. CONWAY, E. J. & O'MALLEY, E. (1942). Biochem. J. 36, BLOC, G. (1948). Acta chem. scand. 2, 467. 655. BREMNER, J. M. (1949). J. Agric. Sci. 39, 183. ELSON, L. A. & MORGAN, W. T. J. (1933). Biochem. J. BREMNER, J. M. (1950). Biochem. J. 47, 538. 27, 1824.

Downloaded from https://www.cambridge.org/core. BBSRC, on 10 Sep 2019 at 09:35:29, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0021859600046232 J. M. BREMNER AND K. SHAW 159 FOLKES, B. F., GRANT, R. A. & JONES, J. K. N. (1950). OGSTON, A. G. & STANIER, J. E. (1950). Biochem. J. 46, J. Chem. Soc. p. 2136. 364. GOTTSCHALK, A. & PARTRIDGE, S. M. (1950o). Biochem. OWEN, O., WINSOR, G. W. & LONG, M. I. E. (1950). J. 46, vi. Nature, Lond., 166, 152. GOTTSCHALK, A. & PARTRIDGE, S. M. (19506). Nature, PARTRIDGE, S. M. (1948). Biochem. J. 42, 238. Lond., 165, 684. QUASTEL, J. H. & SCHOLEFTELD, P. G. (1949). Nature, HOROWITZ, H. N., IKAWA, M. & FLING, M. (1950). Lond., 164, 1068. Arch. Biochem. 25, 226. RAMAAN, E. (1888). Landw. Jb. 17, 405. IMMERS, J. & VASSEUR, E. (1950). Nature, Lond., 165, ROLLER, E. M. & MOKAIG, N. (1939). Soil Sci. 47, 397. 898. SCHOLLENBERGER, C. J. (1930). Soil Sci. 30, 307. IMMERS, J. & VASSEUR, E. (1952). Acta chem. scand. 6, SCHLOSS, B. (1951). Anal. Ohem, 23, 1321. 363. SHINN, M. B. (1941). Industr. Engng Ohem. (Anal, ed.), JOHNS, R. G. S. & MARRACK, J. R. (1952). Biochem. J. 13, 33. 50, xvii. SIDERIS, C. P., YOUNG, H. Y. & KRATJSS, B. H. (1938). JOHNSTON, J. P., OGSTON, A. G. & STANIER, J. E. (1951). J. Biol. Ohem. 126, 233. Analyst, 76, 88. SMITHIES, W. R. (1952). Biochem. J. 51, 259. KOJIMA, R. T. (1947). Soil Sci. 64, 157. SMITHIES, W. R. (1953). Biochem. J. 53, xxix. I/OTWAX-MANN, C. (1941). Biochem. J. 35, 610. TRACEY, M. V. (1951). Biochem. J. 49, xx. MARKHAM, R. (1942). Biochem. J. 36, 790. TRACEY, M. V. (1952). Biochem. J. 52, 265. MEIXAN, I. (1941). Organic Reagents in Inorganic VASSEUR, E. & IMMERS, J. (1949). Arkiv Kemi, 1, Analysis. Philadelphia: Blakiston. 253. MORGAN, W. T. J. (1936). Biochem. J. 30, 909. WAKSMAN, S. A. (1938). Humus, 2nd ed. London: MORGAN, W. T. J. & ELSON, L. A. (1934). Biochem. J. Bailliere, Tindall and Cox. 28, 988. YOSIZAWA, Z. (1950). Tolwku J. exp. Med. 53, 125. MttLLER, P. E. (1887). Studien fiber die naturlichen ZUCKERKANDL, F. & MESSINER-KLEBERMASS, L. (1931). Humusformen. Berlin. Biochem. Z. 236, 19.

{Received 18 April 1953)