Uvb

0.7 ± 0.7

7.9±2.1

5.5 ± 1.4

UVB + Vehicle

1.2 ± 0.8

8.0 ± 5.0

7.6±5.0

UVB + 0.5% aloe

1.2 ± 0.8

8.0 ± 5.0

Values summarize data from Table II in Strickland etal. (1994). There are no statistically significant differences due to treatment effects.

Note

Values summarize data from Table II in Strickland etal. (1994). There are no statistically significant differences due to treatment effects.

Damage to DNA and its repair are crucial events in carcinogenesis, and increased DNA repair is one of the few previously available theraputic interventions for postexposure skin protection. Therefore we investigated whether aloe extracts increased the rate of DNA repair (Strickland etal, 1994). The number of cyclobutyl-pyrimidine dimers in UVB treated skin were quantitated using the endonuclease sensitive-site assay and alkaline agarose gels (Yarosh and Ye, 1990). Irradiation with 5,000-10,000Joules/m2 resulted in 40 to 56 dimers per 106 base pairs, depending on the dose. Even the highest dose of aloe employed (16.7mg/ml) had no effect upon the numbers of cyclobutyl-pyrimidine dimers. Thus the therapeutic action of aloe takes place at a step downstream from DNA damage and repair.

CHS and DTH are differentially protected from UV-induced suppression by process A crude aloe gel

The studies above established that crude aloe extracts can protect the skin immune system from suppression by UV radiation and explored some aspects of its mechanism of action (Strickland etal., 1994). We next sought to dissect the various aspects of the skin immune system using crude materials (Byeon etal, 1998). In particular we asked 'is protection of CHS and DTH responses mediated by the same or different agents in aloe gel?' Differential responses establish that CHS and DTH are quite mechanistically different. Aloe offers the opportunity to determine which response, DTH or CHS, is most important in skin tumor immune surveillance.

One way of accomplishing this with crude material is to determine if the agent protecting CHS has the same stability as the agent protecting DTH. Bioassay data obtained from three lots of crude, lyophilized aloe gel (ARF91A, ARF94B, and ARF94G) were analysed for their stability in protecting CHS and DTH immunity. The responses to antigens vary from experiment to experiment to some degree, thus necessitating the use of the normalizing Formula 1 in order to reduce experiment to experiment variability. When experiments are conducted over a long period of time, the degree to which suppression occurs may also begin to vary significantly. Therefore it is necessary to normalize using the slightly more complicated Formula 2, below, in order to factor out that source of variability.

|m swelling irradiated aloe treated - |m swelling suppressed control

| m swelling positive control - | m swelling suppressed control

Formula 2 Calculation of protection of skin immune response.

The positive control is the aloe-treated, unirradiated pair for each of the UV irradiated, aloe-treated groups. The suppressed control consists of UV exposed animals either without treatment or treated with control vehicle. This process of double normalization to both positive and negative controls allows comparisons between experiments separated by some time with minimal experiment to experiment variability.

In groups of UV irradiated mice treated with phosphate buffered saline (PBS) alone, a 50-80% reduction in CHS and DTH responses is observed compared to their unexposed, matching controls. This degree of response represents 0% restoration while the response of unirradiated, aloe-treated sensitized positive controls was set as 100%.

The results presented in Figure 12.1, show that treatment of UV-irradiated skin with aloe extract prevented suppression of the CHS response to hapten to a variable degree.

Figure 12.1 Stability of CHS and DTH-protective activities in aloe gel extract.

The effect of Process A A. barbadensis gel extracts on UV-induced suppression of contact and delayed type hypersensitivity responses was measured using three lots (ARF91A, ARF94B and ARF94G) of Process 'A' gel at various times after manufacture. For CHS assays ventral skin of groups of five mice was exposed to 2 kJ/m UVB, the irradiated skin was treated with 5 mg/ml gel or saline, three days later, UV-irradiated and unirra-diated control mice were sensitized with hapten and challenged five days later. For the DTH response dorsal skin of mice was exposed to 5 kJ/m UVB radiation, treated with aloe or PBS, UV-irradiated and unirradiated control animals were sensitized three days later with 2 X 107 formalin fixed Candida albicans cells s.c. and their DTH response was elicited ten days later. The aloe extracts were stored as a lyophilized powder at -20 °C until use. The data are normalized according to Formula 1 and are the mean ± SEM. Modified from Figure 1 of Byeon etal. (1998).

The protection afforded by the three different lots of gel varied. For example, three months after manufacture, application of ARF91G gel to UV-irradiated animals provided only 43% restoration of their CHS response while the ARF91A lot of gel extract completely restored the CHS response. The activity of ARF91B was intermediate between these two values. The levels of protection provided by the gel were maximal at a dose of 5 mg/ml (w:v) and could not be improved by increasing the dose used (not shown). The activity of all three lots of gel extract decayed with time, despite their storage as a lyophilized powder. After nine months, none of the extracts prevented UV-induced suppression of CHS responses to hapten (Figure 12.1). Other lots of 'Process A' gel extract gave similar results except that the levels of CHS protection ranged from 30 to 100% and longevity of CHS-protective activity ranged from three to nine months after manufacture. Commercially prepared (non-Process A) gels from the same source were uniformly inactive even when tested within one month after manufacture.

Systemic suppression of DTH responses to C. albicans was measured in mice whose shaved dorsal skin was exposed to 5 kJ/m2 of UVB and treated with the same lot of gel extract used for CHS-protection studies above. In contrast to the partial protection of the CHS responses to hapten, the gel completely prevented systemic suppression of DTH to C. albicans. The immunoprotective activity of all three lots of lyophilized extract remained unchanged after 12 months of storage (Figure 12.1). At each time point at which the gel was tested, the experiment was repeated at least once in order to confirm the results. Restoration of immunity was not due to non-specific immunostimulation by the gel, since neither CHS nor DTH responses were affected in unirradiated control animals at any dose tested (Strickland etal., 1994,). Also, the structurally unrelated polysaccharide, methycellulose, failed to protect CHS and DTH responses against UV-induced suppression.

The different decay rates observed for CHS and DTH-protective activity in crude gel suggested that these activities are mediated by different factors. Additional evidence for the presence of distinct factors was obtained by performing a dose response. From 0 ^g to 5,000 ^g aloe gel in PBS was applied to unirradiated controls and to the UV-irradiated skin of mice immediately after exposure. Three days later the animals were sensitized with hapten through UV-irradiated skin (local CHS model) or injected with formalin-fixed C. albicans cells (systemic DTH model). The data, presented in Figure 12.2 show that protection of CHS responses against suppression by UV radiation was mediated by only the highest dose (5,000 ^g) of gel. In contrast, as little as 10 ^g gel completely protected DTH responses.

These experiments were performed using ARF Standard aloe gel four months after manufacture. The levels of protection were consistent with those observed in Figure 12.1 for gel of that age. Similar results were observed using lots ARF 94G, ARF 94B of crude gel, and an oligosaccharide-enriched fraction prepared from cellulase-treated ARF 94G aloe polysaccharide by hollow fiber ultrafiltration (see Section IIIB). Taken together, the results indicate that protection of CHS and DTH immune responses from suppression by UV radiation is mediated by at least two separate factors in crude aloe gel.

Figure 12.2 Protection of CHS and DTH responses from suppression by UV radiation requires different doses of Aloe barbadensis gel.

Groups of five mice were exposed to 2 kJ/m2 (CHS) or 5 kJ/m2 (DTH) UVB radiation followed immediately by topical applicaiton of 1 |lg to 5 mg of aloe gel in saline. Three days later the mice were sensitized with DNFB or C. albicans and challenged. The data, from three experiments using five animals each are expressed as percent protection using Formula 2 and are the Mean±SEM. Modified from Figure 1 of Byeon etal. (1998).

Figure 12.2 Protection of CHS and DTH responses from suppression by UV radiation requires different doses of Aloe barbadensis gel.

Groups of five mice were exposed to 2 kJ/m2 (CHS) or 5 kJ/m2 (DTH) UVB radiation followed immediately by topical applicaiton of 1 |lg to 5 mg of aloe gel in saline. Three days later the mice were sensitized with DNFB or C. albicans and challenged. The data, from three experiments using five animals each are expressed as percent protection using Formula 2 and are the Mean±SEM. Modified from Figure 1 of Byeon etal. (1998).

Conclusions from studies with crude aloe gel

Crude, Process A, ARF gel extract was capable of protecting the epidermal immune DTH and CHS systems from UVB-induced damage. This protection does not involve absorbing the UV radiation in a sunscreen-like manner or by altering DNA damage or repair. Although these lots of Process A crude gel had the anti-inflammatory activity (R.H. Davis, unpublished observations) classically ascribed to it (Davis etal., 1986, 1987), this anti-inflammatory activity did not prevent the UVB-induced edema of sunburn. The cytoprotective activities and phagocyte activating activities in these materials were complex and will be presented later. The likeliest step, therefore, at which aloe extracts act is the immune cytokine cascade (reviewed below). However, before further applications could be pursued or mechanisms explored it was necessary to characterize the active ingredient.

THE STRUCTURE OF ISOLATED, PURIFIED POLYSACCHARIDES FROM VARIOUS SPECIES OF ALOE AND THEIR PROBABLE STRUCTURE IN THE PLANT

In the literature on biological activity and quality control of aloe materials many investigators insist that the molecule they are studying is the only biologically active molecule in aloe. Organic chemists tend to focus on the anthraquinones and chromones, biochemists on the proteins, and immunologists on the polysaccharides. The authors and our associates are indebted to Robert H. Davis, a physiologist working for many years on Aloe barbadensis extracts and applying countless biological assays to crude extracts, who was, perhaps, the first to suggest that what we were observing in aloe was the combination of several different active moieties of differing chemical structure, operating with different mechanisms and combining to produce a given effect. Davis' paradigm has certainly proved to be true in the case of the effect of aloe on the interaction of UVB and the skin immune system. We suspect that in this system three factors are interactively operating, all chemically distinct and all probably using three different mechanisms. Here we will discuss only one group of molecules, glucomannan (see Figure 12.3), the family comprising the native aloe polysaccharide, and the two fragments that are cleaved from it, the epithelial-protective oligosaccharide and the macrophage-activating acetylated mannan. This saccharide focus does not mean that the cytoprotective oligosaccharide which we will subsequently describe, is the only biologically active molecule in A. barbadensis nor is it the only one with effect on skin. Similarly, the macrophage-activating acetylated glucomannan polysaccharide we will immediately describe does not subsume all aloe biological activity. We have simply chosen at present to focus on this family of interesting molecules. There may be at least two other non-saccharide families of molecules which are also important in protecting the skin against environmental injury.

Aloe barbadensis polysaccharides

There are at least three polysaccharides readily apparent in commercially feasible extracts of A. barbadensis, an acetylated glucomannan, a galactan, and a pectin (for review see Pelley and Strickland, 2000). The dominant polysaccharide of aloe gel is the glucomannan

Biologically inactive Native Aloe Polysaccharide Gel

Biologically inactive Native Aloe Polysaccharide Gel

aJUI I

Low (3 |ig/100 mg) Concentrations of ß 1,4 endoglucosidase aJUI I

'Cellulase Sensitive1 Linkages

Low (3 |ig/100 mg) Concentrations of ß 1,4 endoglucosidase

'Cellulase Sensitive1 Linkages

'Acemannan' Biologically Active in Macrophage Activation Alcohol insoluble

Cytoprotective Oligosaccharide Biologically Active in Epithelium Protection Alcohol Soluble

'Acemannan' Biologically Active in Macrophage Activation Alcohol insoluble

Cytoprotective Oligosaccharide Biologically Active in Epithelium Protection Alcohol Soluble

High (30 mg/100 mg) Concentration of Enzyme

Inactive oligosaccharide

(Typical Commercial Product)

'Acemannan' Biologically Active

Figure 12.3 Cartoon of the putative structure of the native polysaccharide of A. barbadensis and its cleavage by ß 1 ^ 4 endoglucosidase. Modified from Sheet 1 of Strickland etal. (1998).

classically described by Roboz and Haagen-Smit (1948). Another glucomannan was also examined although in fairly degraded material, by Segal etal. (1968). Gowda etal. (1979) were the first to determine the structure of this major aloe polysaccharide, purified by dialysis and then alcohol precipitation. These investigators further fractionated the precipitated polysaccharides based on a gradient of alcohol concentrations. They found a glucose to mannose ratio of 1:4.5 in the more lightly acetylated (9.25%m/m), less alcohol-soluble fraction. The more highly alcohol-soluble fractions had a somewhat higher mannose content (glucose:mannose 1:13.5 to 1:19) and were somewhat more heavily acetylated (10.3 to 17.2% m/m). In all cases a molecular weight of >200kDa was assigned based on total exclusion from G-200 chromatographic gel. They describe the material initially isolated as a jelly. After further purification they describe a material that, although of a molecular weight beyond the resolving capacity of their analytical method, was relatively non psuedoplastic, since it could be subjected to column chro-matography. This is the first published instance we have been able to find wherein the investigators noted that the native gel was of an extremely high molecular weight and that with time and purification it broke down to a non-psuedoplastic state.

All three of the above investigators stated that they were studying 'Aloe vera.' Unfortunately, we have no way of absolutely determining exactly what plant they used. The species under investigation is important since there are significant differences in the polysaccharides from different species, first noted by Mandal and Das (1980b). Commercial plantings in the western hemisphere refer to Aloe barbadensis, strictly an outdated name for A. vera, while cultivation of any other species is extremely rare. Leaves described by Roboz and Haagen-Smit (400 g, 45% yield of gel) are consistent in our experience only with our strain A. barbadensis. In the case of Gowda etal. (1979) the source was simply stated as 'locally available Aloe vera and insufficient detail is given of the isolation to allow us to speculate as to the actual species or variety. Mandal and Das (1980b), working in West Bengal (eastern India), commenting on the work of Gowda etal. (1979) working in Mysore (southern India), noted a difference between the material they were working with and the material Gowda etal. (1979) were working with. Mandal and Das noted that these differences were potentially important for the type of polysaccharide isolated and we agree.

Analysis of a glucomannan purified by alcohol precipitation followed by precipitation with Fehling's solution yielded a polysaccharide with a glucose to mannose ratio of 1:20.6 and suggested a repeating subunit of 3.3 kDa (Mandal and Das, 1980a). The average molecular weight determined by osmometry was 15 kDa. This group also purified and characterized two other polysaccharides from A. barbadensis(sic), one of which is a galactan (Mandal and Das, 1980b) and the other of which is a pectin (Mandal etal, 1983). We agree with their findings (Pelley etal, 1998). We find that the glucomannan greatly predominates in the freshly harvested gel, perhaps comprising >90% of all polysaccharide. However, this fresh gel is subject to degradation by P 1-4 endoglucosi-dases, often termed 'cellulase.' Some of this activity appears to be endogenous. It is fairly well known throughout the industry that loss of psuedoplasticity proceeds with time even in gel to which no exogenous 'cellulase' has been added and which is devoid of bacterial or fungal contamination. More often, in commercial materials, the acetylated glucomannan is broken down by glycosidases of proliferating, contaminating microorganisms (Pelley etal, 1993; Waller etal, 1994) or by fungal 'cellulase' exogeneously added to decrease viscosity and thereby increase yields (Coats, 1994). Often the gluco-mannan is broken down to such an extent that the galactan becomes the predominate polysaccharide (Pelley etal, 1998). Thus, it not uncommon for commercial aloe materials to have a polysaccharide content very different from those classically described in the scientific literature (Pelley etal, 1998).

Polysaccharides from other Aloe sp.

Polysaccharides have also been isolated and characterized from other Aloe species (A. arborescens Miller, A. plicatilis(L) Miller, A. vahombe(sic) Decorse et Poisson, A. vanbalenii Pillans and A. saponaria(Kit ) Haw). The first of these to be characterized, in fact the first aloe polysaccharide to be subjected to detailed examination, was the mannan of A. arborescens (Yagi etal, 1977). These investigators dialysed the gel, precipitated the protein with chloroform and isolated the polysaccharide by repeated cycles of precipitation with acetone. Analysis of the isolated polysaccharide by hydrolysis and paper chromatography revealed mannose to be the predominant sugar. Spectroscopy suggested that the sugars were P-linked and saponification revealed that the mannan was acetylated. The molecular weight, determined by ultracentrifugation, was 15 kDa. Lastly, these investigators were the first to determine a biological activity for an aloe polysaccharide. Ten injections of either 5 or 100 mg of A. arborescens mannan produced a 38 to 48% reduction in the growth rate of Sarcoma 180. This is a classical Biological Respose Modifier (BRM) assay — stimulation of immune responses to tumors (Pelley and Strickland, 2000), although as we have noted, not an assay with strong predictive power.

In 1978 Paulsen etal. reported the isolation of a glucomannan from A. plicatilis gel by dialysis. This polysaccharide had a glucose:mannose ratio of 1:2.8 and a molecular weight by gel filtration upon Sepharose 4B of at least 1,200 kDa. Reducing sugar analysis indicated that the polysaccharide was essentially linear and almost all sugars bore at least one acetyl group.

The polysaccharides of Aloe saponaria and A. vanbalenii were isolated by alcohol precipitation and dialysis and chemically characterized by Gowda (1980). The predominant polysaccharide of A. saponaria was a linear mannan with negligible amounts of glucose and a 20.7% (m/m) degree of acetylation. The predominant polysaccharide of A. vanbalenii was also a pure mannan with significant (19.5% m/m) acetylation. Neither molecular weight nor biological activity were reported.

Aloe vahombe has been described as being a BRM (Solar etal, 1979). Radjabi-Nassab etal. (1984) isolated the polysaccharide by ethanol precipitation and gel filtration using Sephadex G-100. The 4 linked polysaccharide, which eluted in the void volume upon gel filtration (molecular weight >100kDa) had a glucose:mannose ratio of 1:2.6 and 33% of the glucose residues were acetylated. In these respects, this polysaccharide more closely resembles that found in A. plicatilis than it does A. barbadensis. Further studies of the A. vahombe polysaccharide by Radjabi-Nassab etal. (1984) confirmed that the basic 4 linkage was of the P configuration and that cellulobiose-like units were rare.

More recently, two mannans from A. saponaria were isolated by dialysis, size exclusion chromatography and ion exchange chromatography (Yagi, 1984). A. saponaria (As) mannan-1, isolated from material harvested in September 1980 consisted of a linear 4 linked, acetylated (18% m/m) polysaccharide composed exclusively of mannose residues. Molecular weight based on gel permeation chromatography was 15 kDa. As-mannan-2 isolated from material harvested in December 1980 was similar to mannan-1 excepting that its molecular weight was 66 kDa, a 'trace' amount of glucose was present, the degree of acetylation was lower (10% m/m) and branching was evident. Biological activity of mannan-1 after parenteral administration was evidenced by inhibition of carrageenin-induced edema. Later, neutral polysaccharide was isolated by dialysis, chromatography on DEAE cellulofine and gel permeation on Sepharose 6B (Yagi, 1986). Three polysaccharides were observed which differed in their molecular weight and structure. Some confusion exists in this publication because the positions at which polysaccharides A, B and C elute from the preparative gel filtration column (A largest, B midsized and C smallest) do not correspond to the molecular weights described for the polysaccharides in the text (A, MW 15 kDa; B, 30kDa; and C, 40kDa). Assuming that the assignments in the text are correct and that Figure 2 is mislabelled, the conclusions are as follows.

Polysaccharide A (MW 15 kDa) is present in the largest quantity and it is dextran-like (1 ^6 linked glucose). Polysaccharide C (MW 40 kDa) is present in second largest amounts and it is an acetylated (10% m/m) mannan of P1 ^ 4 linkage. Present in trace amounts was an arabinogalactan (Polysaccharide B) of intermediate (30 kDa) molecular weight. Polysaccharide C, which corresponds to the polysaccharide described in 1977 with the exception of its molecular weight, 15 kDa in 1977 and 40 kDa in 1986, was biologically active in that it enhanced phagocytosis and promoted the reduction of NBT dye by human leukocytes. This finding suggested that their polysaccharide was a classical BRM having both anti-tumor effect, described in 1977 and activating effect on phagocytes.

Conclusions about the P-glucomannans of Aloe sp.

Findings over the period 1948-1986 established the structure and biological activity of some aloe polysaccharides. They are variably-acetylated, predominantly-linear polymers of mannose with a P1 ^ 4 linkage. In some cases, they have a significant glucose content. There appear to be two forms described. One is a highly linear mannose-rich form of molecular weight ~15 kDa and the other is of higher molecular weight, generally indeterminant because of technological limitations perhaps related to degradation. These polysaccharides are immunostimulants upon parenteral injection, acting by phagocyte activation. The parallels between this system and the immunostimulatory P1 ^ 3 glucans (Ross etal, 1999; Pelley and Strickland, 2000) which use the CD11b/CD18 receptor are striking. Usually when considering receptor-ligand specificity, the difference between a P1 ^ 3 linkage and P1 ^ 4 linkage is so great that one would not even consider the possibility of cross-reactivity. None the less, given the similarities between the two systems it will be interesting to see if the acetylated glucomannans of the various Aloe sp. use the Mac(macrophage)-1, (CD11b/CD18) receptor.

Aloe polysaccharides, Acemannan®, Carrisyn® and the patent literature

McAnalley at Carrington Laboratories used the findings of the late 1970s and early 1980s to develop an industrial-scale process to isolate A. barbadensis polysaccharide using alcohol precipitation. Recently we have reviewed the history of their Acemannan® and Carrisyn® products (Pelley and Strickland, 2000). The remaining legacy of that story is the numerous polysaccharide patents that have been filed. The shortcomings of the two series of Carrington Laboratories' patents can be addressed (see McAnalley, 1988; McAnalley etal, 1995 and attached bibliographic notes). First, it was hasty to claim the P1 ^ 4 glucomanan structure since it had been previously well described in the literature as we have seen above. Second, the biological activities were well established in the industry and the BRM nature of the polysaccharides were taught as prior art by numerous scientists, cited above. Third, they ascribed all biological activity in aloe to Acemannan® and Carrisyn® polysaccharide. Fourth, the primary analytical tool they employed, FTIR, is extremely inefficient in establishing the purity of complex carbohydrates. Therefore, they consistently overestimate the purity of their materials and the process they developed produced only a crude product. Lastly, although they did teach that the freshness of the raw material was important and breakdown of polysaccharide was to be avoided, they failed to realize that the material they were starting with was already highly broken down. Thus they misappraised the structure of the A. barbadensis glucomannan polysaccharide as it exists in the living plant. As we subsequently successfully pointed out (Strickland etal., 1998), the glucomannan is not a homogeneous linear polysaccharide of about a million molecular weight but, rather, a block copolymer gel.

McAnalley and Danhoff implied that anything which precipitated from crude aloe extracts with alcohol is Acemannan® polysaccharide. In reality, aloe extracts are plant juices and contain very significant amounts of mixed salts of sodium/potassium/calcium malate or oxalate which precipitate with alcohol. Depending on the type of aloe product examined, these mixed organic acid precipitates may comprise up to 50% of the mass of the precipitated 'polysaccharide' (Pelley etal., 1998). This is why in the literature discussed above, polysaccharide is generally purified by dialysis first and then alcohol precipitation afterwards, occasionally purified by precipitation first and dialysis later but never purified by precipitation with alcohol alone. The consequence of this misunderstanding was that many in the industry used the 'Methanol-Precipitable Solids' method as a specific test to measure 'Aloe vera' polysaccharide. Maltodextrin, indeed, was offered as 'Aloe vera' before proper analytical chemistry put a stop to this fraud, adulteration and misrepresentation (Pelley etal., 1998).

'Acemannan' structure and molecular weight by process and analysis

The molecular weight and the structure of the saccharide lies at the heart of the aloe polysaccharide problem. The problem arises because the polysaccharide was never purified to chemical homogeneity, and the appropriate molecular weight analysis performed before biological testing. In U.S. Patent No. 4,735,935 patent McAnalley claims that all acetylated $1 ^ 4 mannans from the disaccharide (degree of polymerization n = 2) to the high linear polymers (n = 50,000) were found and that the polymer is at least 80% mannose. This implies a linear structure, while the native polysaccharide is obviously a gel.

In the examples given in the various patents the molecular weights of the various materials exemplified vary greatly. By way of process, in the '935, patent, Example 1 (columns 20—21) specifies that the product is produced by ultrafiltration, which removes undesirable compounds of less than 10 kDa (nominal) and retains the polysac-charide, presumably of nominal molecular weight >10kDa. It is further specified that this retained fraction can be further fractionated by passage through an ultrafilter of nominal molecular weight cutoff 50 kDa wherein desired material is obtained. This would define Acemannan® as having a molecular weight of 10—50kDa. Example 27 of the '935 patent illustrates molecular weight as determined by high pressure liquid chromatography using a 7.5 X300mm Beckman Spherogel TSK 2000 column. Unfortunately, detection of eluted material was by nonspecific (refractometry) methods so that the peaks are not defined. Furthermore, the chromatograms, which are crucial to evaluating the data, are not shown. From the tabulation of calibration it is apparent that analytical precision is obtained only in the molecular weight range of 40 to 9 kDa. Three classes of materials were exemplified with molecular weights of >80 kDa, >10kDa and < 1 kDa. From the summary (65, lines 28—30; 'cleave the function groups and glycosidic bonds ($(1 ^4)) thus reducing or eliminating its activity.') and Claim 1a ('substantially non-degradable'), it is apparent that materials in the third chromato-graphic region are undesirable. By reference to the published scientific literature reviewed above, it is highly probable that the first described materials (50, line 59

Table 12.11 Molecular weight distribution of Acemannan®.

Percentage of material in class (range)

Fraction #1 Fraction #2 Fraction #3 (>80 kDa) (>10 kDa) (undesired)

Experimental 47%

Manufactured 28%

' Fraction #1, MW>80kDa.') corresponds to the material described by Gowda etal. (1979) and that the second material (line 60, MW>10kDa.) corresponds to the material of molecular weight 15 kDa described by Mandal and Das (1980). The third material consists of undesired material of molecular weight < 10 kDa. This undesirable low molecular weight material was removed by dialysis or by solvent precipitation of the desired higher molecular weight polysaccharide, in the investigations by Gowda etal. (1979) and Mandal and Das (1980) immediately above. By these examples, Acemannan® could be a 40 kDa linear polysaccharide or a >80 kDa polysaccharide.

With all of these assumptions in mind, the examples of McAnalley can be understood as follows. There are two polysaccharides in Acemannan®. One is of molecular weight greater than 80 kDa, although the actual molecular weight cannot be more precisely determined because it is beyond the analytical range of the method. The second polysaccharide is of molecular weight of perhaps 12 kDa. There is a third component of undesired lower molecular weight contaminants and breakdown products. The content in Acemannan® of these materials varied greatly in Example 27 of the '935 patent as is seen in Table 12.11 below.

As interpreted by McAnalley this data implies that the process described is relatively uncontrolled in its enzymatic breakdown until a subunit size of ~10—15 kDa is reached after which breakdown occurs to very small fragments. Our findings are in agreement with this. It is then infered in Example 27, Example 31 and Example 32 that the process of monitoring the breakdown of Acemannan® can be achieved by IR spectroscopy. However, this method cannot measure the molecular weight distribution. What appears to be happening is that the polysaccharide is being broken down prior to alcohol precipitation. The <5 kDa saccharide fragments are remaining in the supernatant but the mixed mineral cation/organic acid coprecipitate is still precipitating with what little polysaccharide is left. This means that the resulting FTIR pattern shows that the Acemannan® product is increasingly contaminated with alcohol-insoluble malate and oxalate salts. The same phenomenon can be observed with broken-down commercial material. Example 5 of US Patent 4,851,224 is concerned with, among other things, the isolation, purification and characterization of the Carrisyn® Acemannan® polysac-charide. This section of the 224 patent has errors similar to those referred to above. With these in mind, the data can be interpreted as being consistent with the Acemannan® product consisting of a Material 'A' of molecular weight >100,000 and a Material 'B' having a molecular weight 'greater than 10,000 but less than 100,000 daltons.' Inspection of the figures, in fact, suggest a molecular weight of approximately 12 kDa for Material 'B'. It is further stated that ' the sum of peaks A and B constitute the active fractions' although the nature of the activity is unspecified and unexemplified. This data, taken together with the Summary (42, lines 21—31) indicates that 'A' (>100kDa) is active and upon decomposition converts to 'B' (~12kDa) and subsequently to inactive 'C' (dialysable, alcohol-soluble, MW<10kDa). The conclusion is that the Acemannan® is breaking down to a ~12 kDa fragment and from there to inactive fragments, but it is not known from what it is breaking down. Furthermore the assumption is that breakdown is to a single entity, fragments of the acetylated mannan.

Consensus summary of polysaccharide structure

There appears to be a consensus between the publications in the scientific literature and the patents reviewed above, concerning the structure of aloe polysaccharides. There exists in aloe an essentially linear, acetylated polysaccharide of discrete molecular weight between 12 kDa and 15 kDa. In the case of A. arborescens (Yagi etal, 1977), A. barbadensis (Mandal and Das, 1980) and A. saponaria (Yagi etal, 1984) it has the following structure:-

(2,3-acetyl mannose, ^4, 6-acetyl mannose)28-35

Structure 1 The acetylated mannan disaccharide repeating unit (MW 450) and its polymeric polysaccharide, Acemannan®. With a polymer molecular weight to 12.6 to 15.7 kDa, there would be approximately 28 to 35 repeating units.

This is the structure later claimed by McAnalley (1988). There also exists a higher polymer of this structure, possibly linked together through 1,6 linkages:-

[(2,3-acetyl mannose, ßl ^4, 6-acetyl mannose)28_35]n

Structure 2 The polysaccharide formed by linking together Acemannan® units into a higher molecular weight form but not crosslinked into a gel, n ~8.

where n is at least 8 and probably generally much larger. This structure 2 has been reported for A. barbadensis (Gowda, 1979), A. saponaria (Yagi etal, 1984) and A. arborescens (Yagi etal, 1986) and also claimed by McAnalley (1988). Mannans of chemical composition similar to the above have been described for A. saponaria and A. vanbalenii (Gowda, 1980) although it was not stated whether they were present as Structure 1 or Structure 2. Glucomannans have been described for A. plicatilis (Paulsen etal, 1978) and A. vahombe (Radjabi etal, 1983) with structures similar to Structure 2 excepting that glucose substitutes for some of the mannose residues.

A novel model for native polysaccharide structure

Structure 2, however, fails to account for three physico-chemical findings. First, how do we account for the 5 to 10% glucose that is so consistently found in the major A. barbadensis polysaccharide? Second, how do we account for the psuedoplastic gel observed at such an extremely low (0.5 g%) concentration? Third, why is there such sensitivity of this gel to the viscosity-reducing activity of 'cellulase,' given the relative resistance of acetylated mannan to the reducing sugar generating ability of this enzyme? These shortcomings of Structure 2 were fairly easy to ignore because almost no investigators had access to Process A type material. Rather they were working with material that mostly had the composition of Structure 1 or low multimers thereof.

Fractionation of the ARF Process A gel revealed its unusual physical properties and the correlation of changes in these properties with biological activity. We have claimed (Strickland etal., 1998) that the structure (Figure 12.3) of the native glucomannan gel from A. barbadensis consists of linear portions of Structure 1, above, crosslinked into a gel by oligosaccharide linkers. We believe that these linking oligosaccharides may have 1,4,6 mannose branch points and that the P1 ^ 4 glycopyranosides adjacent to the branch points are unacetylated and thereby considerably more cellulase-sensitive than Acemannan®. Thus, cleavage proceeds most rapidly by clipping adjacent to the branch points and liberating linear acetylated mannan polysaccharides of molecular weight ~10-15 kDa, branched unacetylated oligosaccharides of molecular weight <5 kDa, and mixtures of the two.

Biological activity for suppressing inflammation and increasing the immune response has been claimed for both Structure 1 and Structure 2 by Yagi etal. (1977, 1984, 1986) by activating phagocytic cells. McAnalley (1988) claims an increase of the rate of wound healing by increasing the growth of fibroblasts and phagocytic activity. We agree that acetylated polysaccharide of molecular weight >10 kDa is capable of activating phagocytic cells (Strickland etal., 1998, Figure 9). However, we do not think that the wound-healing and epithelial cell stimulating activity in aloe is due to a linear acetylated P1 ^ 4 mannan. We have demonstrated that A. barbadensis oligosaccharides, <5 kDa in molecular weight, are protective from UV radiation of the skin immune system. We suspect that the weak fibroblast activity noted by McAnalley above is due to a cytoprotective oligosaccharide contaminating his impure materials. The family of cytoprotective oligosacchrides appears to act, in the case of UV, by down-regulating Interleukin-10 production by injured keratinocytes (Byeon etal, 1998; Strickland, 1999).

Was this article helpful?

0 0
Aloe and Your Health

Aloe and Your Health

This іѕ а health guide, whісh саn hеlр people knоw hоw tо mаkе uѕе оf Aloe vera tо improve оvеrаll health аnd deal wіth mаnу kinds оf diseases.

Get My Free Ebook


Post a comment