The Pigmentation of Cavernicolous Animals: II. Carotenoid Pigments in the Cave Environment

The present paper records the results of some analyses of carotenoid material in cave detritus, carried out as part of a research on the causes of pigment lack in the cave fauna. There is reason to suppose that the ancestral pigment of the numerous unpigmented cavernicolous Crustacea was the carotenoid pigment astacin, since closely related above-ground species frequently contain this pigment. Thus the amphipods Orchestia gammarellus and Gammarus marinus have a wholly carotenoid pigmentation of astacin type, while Gammarus pulex is believed to contain astacin (Sörensen, 1936). It is a well-known fact that the pigment of many Amphipoda is soluble in the alcohol in which the animals are preserved, and we are justified in concluding that astacin-like pigments are typical of the group. It is unlikely that Crustacea can synthesize carotenoid pigment from non-carotenoid food (vide the discussion in Verne, 1926), and the present work forms a preliminary survey of the extent to which carotenoid-containing food is available to the cave fauna.

Several modes of entry of food into caves have been noted. The excreta of bats, on which colonies of coprophagous beetles live, is of minor importance. Accidental entry of leaves, twigs, etc., at the mouth of the cave, does not contribute to the food of animals in the main region of the cave. Detritus carried in by underground rivers is the chief source of food for cave animals, and as shown here, carotenoid material is found in it in some quantity. Proteus, which is associated with subterranean rivers, was found to contain some carotenoid material. In the upper regions of caves drip pools are found, fed only by water percolating through the limestone. Detritus from such pools was found to be devoid of carotenoid material, and specimens of the cavernicolous amphipod Niphargus from such a pool were found to contain no carotenoid pigment.

Samples were extracted with acetone, methanol, and in some cases subsequently with ether, or petrol ether (b.p. 50–60° C.). Analysis for carotenoid pigments was carried out on the basis of the usual micro-technique (Kuhn & Brockmann, 1932). The extracted pigment, dissolved in petrol ether, was saponified by adding an equal volume of 5 % potash in absolute alcohol and maintaining the mixture at 40°C. for at least 3 hr. Pigment was regained in petrol ether by the addition of water, and partitioned between petrol ether and 90 % methanol, the carotene remaining in the (epiphasic) petrol ether and the freed xanthophylls passing into the (hypophasic) alcohol layer. The carotene and xanthophyll fractions were transferred to pure petrol ether and then subjected to chromatographic analysis by adsorption on columns of adsorbent material about 1 cm. in width and 8 cm. in height. The resultant coloured bands were eluted and their absorption maxima determined for comparison with the absorption maxima of known carotenoid pigments. As an additional check, the carotenoid nature of the pigments was usually confirmed by treatment of the dry chloroform solution of pigment with a solution of antimoriy trichloride in chloroform, a blue colour being consonant with the presence of carotenoid.

In the experimental section below, the use of the terms ‘epiphasic’ and ‘hypo-phasic’ without further qualification refers to partition between petrol ether (b.p. 50–60° C.) and 90% methanol.

Some mud from the top half-inch of a deposit on the floor of a gallery in Chislehurst Cave, Kent, was collected. The mud appeared to have been washed in from a hole in the roof and to have reached this gallery, which was in total darkness, in suspension in water. The sample was examined exhaustively for carotenoids but none could be detected.

The sample, weight 500 g., was extracted exhaustively with acetone and other solvents, and all pigment transferred to 1000 c.c. of ether, a deep red-brown solution resulting. It was observed during the extraction that all solutions had a strong green-blue fluorescence, that in none of them was an absorption band due to chlorophyll observed in the spectrum, and that when solutions were taken to dryness a red-black solid remained.

A portion was chromatographed in solution in petrol ether on a column of magnesium oxide. Two deep brownish diffuse bands appeared at the top of the column, and could not be eluted by the usual elution media for carotenoids, e.g. petrol ether containing methanol or benzene. After elution with pyridine, a blood-red solution resulted, but the pigment had only a single absorption band and responded negatively to the antimony trichloride test. Chromatographing on magnesium oxide serves therefore to remove a great part of the non-carotenoid pigments, in the extract.

Of the main extract 50 c.c. were examined by the usual micro-method (Kuhn & Brockmann, 1932), and separated into carotene, xanthophyll, and xanthophyll-ester fractions. At no stage was pigment observed to be sharply epiphasic or hypo-phasic. The carotene fraction when chromatographed on magnesium oxide gave rise to a dirty yellow staining at the top of the column. A small yellow band was separated from this and eluted through the column by passage of petrol ether + 2% methanol. Another was eluted by passage of pure benzene. The first eluted fraction exhibited an absorption maximum at 456μμ, the second at.459μμ. Neither fraction gave a positive reaction to the antimony trichloride test. The xanthophyll fraction was not appreciably adsorbed when chromatographed in petrol ether on calcium carbonate. The pigment eluted through the column by petrol ether and benzene 1 : 1 showed an absorption maximum at 456μμ, in carbon disulphide, and gave a negative antimony reaction. The xanthophyll ester fraction was chromatographed in petrol ether on calcium carbonate. Any possible, carotenoid was eluted through the column with petrol éther and benzene 1 : 1 ; the elutrate had an absorption maximum at 457μμ. in carbon disulphide, and failed to respond to the antimony test.

In this sample of detritus therefore no trace of carotenoid or of chlorophyll was observed. Large quantities of a non-carotenoid pigment were present, with the following properties. The pigment is fluorescent in solution, and when dry is a red-black solid. The pigment is not sharply epiphasic or hypophasic. During various processes the pigment was found to be insoluble in water, and saturated aqueous salt solution, sparingly soluble in petroleum ether, moderately soluble in methanol, and very soluble in ether, acetone, benzene, chloroform, and dichloroethane. The absorption maximum of the pigment in carbon disulphide is 457 ± 2μμ. The pigment has not been identified.

The Postumia Grotte of north Italy, formerly known as the Adelsberg Grotto, is the largest cave system in Europe, and contains 27 km. of passages so far explored. The River Piuca which flows through it floods the upper galleries during the winter and leaves pools behind when it recedes. In these pools are found such characteristic cave animals as Proteus anguineus Laur, and the decapod crustacean Troglo-caris schmidti Dorm. In still higher regions of the cave pools of a different ecological status are found—these are never flooded by the river and derive their water from roof drippings. They contain a smaller number of cave animals, notably Niphargus spp. Plant remains are commonly found in those pools subjected to seasonal flooding, while in the drip pools only a fine clay is usually found.

Four samples of detritus were collected :

• Sample A. Earth from just within the entrance to the Grotto Nera. Fauna, various epigean species.

• Sample B. Mud from a small pool in the Gnomo della Grotte, well within the cave. Water in the pool derived solely from roof drippings. Nine specimens of Niphargus spp. collected from the pool.

• Sample C. Debris, chiefly old leaves and pieces of wood, from the edge of the water at the end of a mud passage in the Grotto Nera, well within the cave. Fauna planarians, Proteus, Troglocaris, and Asellus aquaticus cavernicolas.

• Sample D. Mud from the passage described under sample C, but above the water level. The area is flooded in winter. Fauna Titanethes (an amphibious isopod).

75 g. of fresh sample were extracted with acetone and all pigment transferred to petrol ether, 200 c.c. of a fine yellow-brown solution with a greenish fluorescence resulting. Absorption bands approximating to those of chlorophyll were observed in the spectrum.

The entire petroleum ether solution was chromatographed on magnesium oxide. The following system of bands appeared :

A fine yellow solution was eluted through the column by passage of 2% methanol in petrol ether (‘Methanol fraction’). A further quantity of pigment was eluted by passage of 2% benzene in petrol ether (‘benzene fraction’). Various brown and green bands remaining on the column were rejected as being obviously of non-carotenoid nature.

The ‘methanol fraction’ was hydrolysed and partitioned. The epiphase gave a strongly adsorbed yellow band when chromatographed on magnesium oxide in solution in petrol ether. Two small pigment fractions were eluted through the column by successive passage of 0·2 and of 2 % methanol in petrol ether. The two fractions were epiphasic to 90 and 95 % methanol, had a yellow fluorescence, showed no absorption bands in chloroform, and responded negatively to the antimony trichloride test. The hypophase was transferred to dichloroethane and chromatographed on magnesium oxide. Three yellow bands appeared and were eluted separately. The upper band was epiphasic, and probably represented some carotene not fully extracted by the partition with petrol ether. The middle band was not sharply epiphasic. Neither band contained sufficient pigment for further analysis. The lower band was chiefly hypophasic ; the spectrum showed absorption maxima at 508 and 480μ μ in carbon disulphide. A xanthophyll was therefore present.

The ‘benzene fraction’ was saponified and chromatographed in petroleum ether on calcium carbonate. A yellow stain appeared throughout the column, and a brown band at the top. The two regions were eluted separately. The pigment from the brown band responded negatively to the antimony trichloride test and exhibited no absorption bands. The pigment from the yellow staining over the column was almost wholly epiphasic to 90 and 95 % methanol. In chloroform, the spectrum showed sharp absorption bands at 503 and 473 μ μ.

At least two carotenoid pigments were therefore present in small quantity in sample A, as summarized in the previous table. The maxima recorded for γ-carotene are from Winterstein (1933), and those for lutein from Kuhn et al. (1931).

240 g. of fresh sample were extracted with acetone and the pigment taken up in petrol ether. [During the washing of the petrol ether with water, much brownish non-carotenoid matter separated. It was readily soluble to give a deep red solution in ether.] The extract was chromatographed in petrol ether solution on magnesium oxide. A variety of orange, yellow, and brown bands appeared. All possible carotenoid pigment was eluted through the column with 2 % methanol in petrol ether and 2% benzene in petrol ether. The remaining brownish bands, non carotenoid in nature, were discarded.

The elutrate was hydrolysed and partitioned. A deep yellow epiphase and a pale yellow hypophase resulted. The epiphase was chromatographed in solution in petrol ether on magnesium oxide and developed with 0·5% methanol in petrol ethen. A brown band was left at the top of the column and discarded. An orange band below it was eluted (absorption maximum in carbon disulphide 453μμ).

The third and lowest band was diffuse yellow, with an absorption maximum at 456μμ. Neither of the two lower bands responded positively to the antimony trichloride test. The hypophase when chromatographed in solution in petrol ether on calcium carbonate gave only a diffuse yellow staining. The pigment when eluted had an absorption maximum at 455μμ in carbon disulphide, and responded negatively to the antimony trichloride test.

In this relatively large sample therefore carotenoid pigments and chlorophyll were absent. Further, the nine specimens of Niphargus collected from the water yielded no trace of coloration when extracted in the usual way, even when any possible pigment was concentrated in a minute volume of petrol ether.

For comparative purposes, small quantities (30 g.) of each sample were examined under comparable conditions. They were extracted exhaustively with acetone, ethanol and ether, and pigment transferred to ether. The ether solutions were dried over anhydrous sodium sulphate, evaporated to dryness under reduced pressure, and the last traces of solvent removed by heating at 70° in a current of coal gas. The residues were weighed, redissolved in ether, and the spectra examined for chlorophyll bands. Finally each ether solution was hydrolysed and analysed for carotenoid material. The residues of extracted detritus were weighed and cooled in a desiccator to constant weight over a period of months. The properties of the samples are compared in the following table :

A large proportion of the non-carotenoid colouring matter was removed by chromatographing the hydrolysed pigments in solution in petrol ether on magnesium oxide, the carotenoid pigments being recovered by elution of the column with 2% methanol in petrol ether. The eluted carotenoids were hydrolysed anew to saponify any remaining chlorophyll or xanthophyll ester, and were examined for carotenoids by the usual Kuhn and Brockmann microtechnique, with the following results.

Sample A. Various pigment fractions were isolated from the small quantity of sample A used, but none had sharp absorption bands or gave a positive antimony trichloride test. It will be recalled, however, that the investigation of a greater quantity of sample A already described had demonstrated the presence of a small quantity of carotene and xanthophyll.

Sample B. As already described, an analysis of 240 g. of this sample showed no trace of carotenoids.

Sample C. After partition, the epiphase had absorption bands at 512 and 485μμ in carbon disulphide. This approximates to the values of 517 and 486μμ recorded for β-carotene in this solvent by Smith (1936). This pigment and a stock sample of β-carotene gave the same behaviour on a magnesium oxide column when chromatographed in solution in petroleum ether and developed with 2,% methanol in petrol ether. The hypophase gave rise to four yellow bands when chromatographed in petrol ether on a column of calcium carbonate. The upper two bands, indistinctly separate, were eluted together, and the mixture exhibited absorption maxima at 490 and 457μμ in chloroform. The next lowest band showed after elution maxima at 488 and 456μμ in chloroform. The lowest band was large, and exhibited absorption maxima at 492 and 458μμ, in chloroform, and 505 and 485μμ in carbon disulphide. The absorption maxima for the pigment of the lowest band, and its position on the column, suggest a mixture of lutein and zeaxanthin, the two commonest xanthophylls. In actual fact two bands did appear when the pigment of this band was rechromatographed in solution in dichloroethane on a magnesium oxide column. The three fractions eluted from the chromatogram responded positively to the antimony trichloride test.

Sample D. Little pigment was present. It distributed itself during partititon with slightly more pigment in the epiphase than in the hypophase. Insufficient pigment was present to give absorption bands or a decisive antimony trichloride reaction. Probably a little carotene and xanthophyll are present.

The blind and almost colourless cave salamander Proteus anguineus Laur. of north Italy and Yugoslavia is a typical member of the cave fauna. It is always found in association with an underground river, and from the evidence given earlier in this paper it may be assumed to have access to carotenoid-containing food.

Three large entire specimens were extracted by grinding with sand and methanol. Pigment was transferred to petrol ether and separated into carotene, xanthophyll, and xanthophyll ester fractions. The carotene fraction was found to contain the greater proportion of the pigment, and was chromatographed in petrol ether on magnesium oxide. A band of typical β-carotene appearance resulted, which on elution showed absorption maxima at 518 and 489μμ. This agrees fairly well with the values of 517 and 486μμ, recorded for β-carotene in this solvent by Smith (1936). The xanthophyll fraction gave rise to a small yellow band when chromatographed in petrol ether on calcium carbonate, and two rather indefinite absorption maxima could be seen. No trace of xanthophyll ester could be demonstrated.

Pigment from the eviscerated body of two specimens was found to pass almost wholly into the xanthophyll fraction. Pigment from the liver passed almost wholly into the carotene fraction. The guts after cleaning yielded hardly a trace of pigment.

To summarize: the total quantity of carotenoid pigment in Proteus is small. The whole body contains carotene (probably β-carotene) and free xanthophyll. The eviscerated body contains chiefly xanthophyll, the liver chiefly carotene, while the gut is almost devoid of carotenoid pigment.

In cave mud from both England and Italy large quantities of a non-carotenoid red pigment can be extracted by organic solvents. The pigment has a marked absorption maximum in the spectrum at 457 ± 2μμ. It has not been identified.

The distribution of carotenoid material in caves is correlated with the detritus carried in by subterranean rivers. In the Postumia Grotte, air-borne detritus from qear the entrance contained only a little carotenoid material (sample A). Organic detritus from a pool in connexion with the underground river contained no less than 0·73 % of matter extractable by organic solvents per dry weight of sample, and in the extracted material large quantities of carotenoid material were present, including β-carotene and at least four xanthophylls, together with much chlorophyll (sample C). The temperature of water of caves is as a rule uniformly low (about 10° C.) and has a low oxygen tension: this may account for the preservation of the carotenoids. Fine detritus carried in by the river but exposed to the air by the summer recession of the water contained only a little carotene and xanthophyll (sample D). Sample B on the other hand was from a pool not in connexion with the river, and was found to be devoid of carotenoid material ; specimens of Niphargus from the same pool were also free from carotenoid. In such drip pools it would seem therefore that the water which feeds them is freed of carotenoid as it percolates through the limestone, and that there is no synthesis of carotenoid by bacteria, fungi, or the fauna. Matter extractable by organic solvents was at the low level of 0·04 % dry weight of sample.

Proteus was found to contain a little carotenoid pigment in the body tissues. Since Proteus is a carnivore, its chief article of diet being the cave Crustacea, it may be assumed that there is a transference of carotenoid material from detritus to Crustacea, and thence to Proteus. Cave Crustacea therefore seem to pick up. vegetable carotenoid material when it is available. Since, however, the Crustacea are practically colourless and lack the characteristic pigment of epigean forms, it would appear to be established that the specific animal pigment astacin is lacking whether or not vegetable carotenoid material is available in the diet.

In the organic detritus of sample C a preponderance of xanthophyll was demonstrated. Three xanthophyll fractions were isolated with absorption maxima in chloroform at 490 and 457μμ, 488 and 456μμ, and 492 and 458μμ. These figures recall the findings of Baudisch & von Euler (1934), who found that peat rich in calcium carbonate had a considerable xanthophyll content, the absorption maxima of the hydrolysed pigment in chloroform being at 488 and 455μμ. They have suggested the theory that a natural adsorption of xanthophyll takes place on limestone, vegetable oils acting as solvent, while carotenes are but weakly adsorbed, and in the present work there is certainly a preponderance of xanthophyll over carotene in organic detritus associated with limestone. In deep-sea mud Fox (1937) found a preponderance of xanthophyll, with absorption maxima comparable with those recorded by us and by Baudisch and von Euler.

The above investigations have been carried out during the tenure of a Maintenance Grant from the Department of Scientific and Industrial Research. The work has been supervised by Dr E. H. F. Baldwin, to whom I am indebted for advice. I am especially indebted to Mr R. S. Hawes, B.Sc., for his careful collection and transmission to England of the samples from the Postumia Grotte.