Genetics: 11: 73-82 (Jan. 1926)
THE PRECIPITIN REACTION IN RELATION TO GRAFTING1
F. GREEN
Department of Physiology and of Experimental Medicine,
McGill University, Montreal, Canada Received July 24, 1925

1The expenses of this research were in part defrayed by a grant from the JAMES COOPER FUND of McGILL UNIVERSITY.

TABLE OF CONTENTS

 
PAGE
INTRODUCTION
73
Technique
75
Preparation of the extract
75
Immunization
76
The precipitin reaction
77
Conglutination reaction
78
RESULTS AND CONCLUSIONS
80
LITERATURE CITED
82

INTRODUCTION

Since it is apparent that one of the principal factors for the successful production of grafts, and possibly also for hybridization, is that the plants should be nearly related, and since the relationship among animals has been successfully studied by means of immunity reactions, I have tried to employ these more exact and scientific methods in the investigation of plants. For this purpose, I have used what is known as the "precipitin reaction."

This reaction, discovered by KRAUS in 1897, gave to UHLENHUTH (1900, 1901), WASSERMANN (1903) and NUTTALL (1904) the means of differentiating the various species of animals through their specific proteins. As precipitins are formed with any soluble animal or vegetable protein, they were made use of also in botany. The phytoprecipitins, as they are called in botany, were first studied by KOWARSKY (1901), MAGNUS and FRIEDENTHAL (1906), MAGNUS (1908), GASIS (1908) and others. These authors proved that specific precipitins could be obtained from various plants, but did not attempt a methodical investigation in this field, as had been done with animals.

In 1913 the Botanical School of Kšnigsberg, under the direction of MEZ, started methodical researches for the classification of plants by means of the precipitins. This school, guided by the previous classifications based on the morphological and structural appearances of plants, started to test them by means of this immunological reaction (GOHLKE 1913), and, although the new classification is still in its infancy, they have already examined a large number of plants. MEZ and GOHLKE (1913) have studied the Angiosperms; MEZ and LANGE (1914) the Ranales; MEZ and PREUSS (1914) the Parietales; MEZ and KIRSTEIN (1920) the Gymnosperms; SALTZMANN (1924) has reviewed once more the Ranales and completed the Gnetales; RAEDER (1924). the Adoxaceae, Loranthaceae, Ericaceae, Polygalaceae, Empetraceae and Hypericaceae; GUTTMANN (1924) the Archegoniatae; STEINECKE (1925) the genealogy of the Algae. This work, of the most interesting nature on account of its fundamental character,  is not yet generally acknowledged, but, as RAEDER says, "the genealogy of plants established by means of the sero-diagnostic reactions is gaining more and more recognition."

As the literature seems to be mute on the subject of immunological reaction in relation to grafting among plants, I undertook a study of the behavior of the precipitin reaction in relation to a few plants known to be successfully intergrafted. The external characteristics are not an absolutely reliable rule by which one can judge the degree of the internal and constitutional relationship either for the vegetative or for the sexual affinity between the different species.

The object of this research was to investigate whether plants having been classified in families by their morphological characteristics and generally known to be capable of being intergrafted, would react in the same way to the specific biological reaction of the precipitins.

The plants which I have particularly studied are the following:

(1) Family of the Rutaceae; Genus: Citrus.
  Citrus aurantium (orange).
Citrus limonum (lemon).
Citrus vulgaris (bitter orange).
Citrus decumana (grapefruit).
Citrus nobilis (mandarin orange of China).
(2) Family of the Rosaceae; Sub-family: Pomoideae.
  Pyrus communis (pear).
Pyrus malus (apple).
(3) Family of the Rosaceae; Sub-family: Prunoideae.
 

Prunus domestica (plum).
Prunus armeniaca (apricot).
Prunus persica (peach).
Prunus amygdalus (almond).

(4) Family of the Solanaceae.
 

Solanum  tuberosum (potato).
Lycopersicum esculentum (tomato).

(5) Family of the Chenopodiaceae.
  Beta vulgaris (beet).
(6) Family of the Polygonaceae.
  Rheum officinale (rhubarb).
(7) Family of the Oleaceae.
  Olea europaea (olive).

The last three plants mentioned were used principally as controls.

TECHNIQUE

Preparation of the extract

The extract may be prepared from any part of the plant under investigation. I prepared all the extracts from the seeds, which, as previously noted by others, owing to the small amount of water and the large amount of soluble albumins contained in them, give an all-round greater uniformity of the extracts. The dried seeds, preferably of the same season, are reduced to a fine powder by means of a coffee grinder or, if the seeds arc too large, by means of a chopping machine. The powder is freed from chaff by sifting and 5 grams of the fine flour-like powder is put into a perfectly clean and sterile Erlenmeyer flask with 50 cc of sterile 0.9 percent NaCl solution. The mixture is kept at room temperature for 12 hours and is occasionally vigorously shaken. It is then poured onto 6 layers of sterile gauze and well pressed. The milky filtrate thus obtained is then filtered twice through filter paper, Schleicher and Schull No. 595. A sample of the clear filtrate is now taken and its albumin content is estimated by means of an Esbach albuminometer, using the Tsuchiya reagent (phosphotungstic acid 1.50 gm; hydrochloric acid 5 cc; alcohol 95 percent, 95 cc). The reading was taken after 2 hours and the amount of albumin per liter noted. As a rule, the extracts thus prepared averaged an albumin content of 3 grams per liter. As I intended to work with extracts containing approximately the same amount of albumin, if the extract did not conform to that definite amount of albumin, a correction was necessary. This correction was made, naturally, either by increasing the amount of seed flour or by diluting the extract proportionally. I laid great stress on this special point of technique in my research. The main portion of the filtrate was incubated at 60¡ C for one hour, after which time, it was transferred to another incubator at 37¡C for 24 hours. This fractional sterilization was repeated for 3 consecutive days. Afterwards, the extracts were preserved in the ice-chest until required' for use. No chemicals were used as preservatives. As a rule, all extracts, thus prepared and treated showed a certain amount of precipitate due, probably, to the temperature and the acid reaction of the fluid. Although this coincidence. probably diminishes the contents of active substances, it does not seem to interfere with the efficacy of the extract. Extracts which, notwithstanding all the precautions taken, gave rise to moulds, were discarded.

SALTZMANN (1924) prepared the extracts from dried and powdered leaves and to test the presence of albumin in the extracts he used GUTTERMANN'S method of demonstrating the presence of organic substances, using the Berlin bleu reaction for nitrogen and the nitroprussiate reaction for detecting the presence of sulphur. PREUSS and SALTZMANN have had good results in preparing the extract, the former with 0.1 percent and the latter with 0.5 percent NaOH.

Immunization

The animal used for immunization was the rabbit. Brown male rabbits of an average weight of 3000 grams were chosen. Before and during the process of immunization, they were fed only on carrots.

A precise neutralization of the extract previous to injecting the animal. to be immunized is of capital importance. I used sodium hydroxide 0.1 percent and acetic acid 0.1 percent, using phenolphtalein as indicator.

With all the precautions of asepsis, 10 cc of carefully neutralized extract are injected into the peritoneum of the animal. The intravenous way may also be used but the intraperitoneal is safer, as, given the slow absorption of the material, it causes less reaction. The latter way is imperative, however, when immunizing animals with extracts prepared from seeds containing large amounts of toxic substances like solanin, nicotin, etc., in which case the process of immunization must be carried out with very great care and appropriate dosage. An interval of 4 days elapses between successive injections. The number of injections given averages from 7 to 10. A total quantity of 120 to 140 cc is injected, the first and last 2 injections being of 20 cc each and the intervening injections of 10 cc each. Rabbits, as a rule, bear the injections very well. Although at the beginning of the immunization they show no loss in weight, a loss is noticeable in the last few weeks of immunization.

A preliminary test of the serum of the animal so treated, when made to react with the extract used for immunization diluted in convenient proportions, will show, by the formation of a well-defined precipitate, the titer of the serum. The highest titer of my immune sera, as a rule, was between' 6400 and 12,800. If the immune serum is found to give rise to a satisfactory precipitate, the animal is starved for 24 hours and then sacrificed by collecting the blood from one of the carotids, with all the precautions of asepsis, in a sterile Erlenmeyer. The blood is preserved in the ice-chest until the day after, when it is centrifuged; the supernatant clear serum is pipetted off and transferred to 5 cc glass ampoules, sealed at the dame without adding any preservatives and put in a dark, cool place.

The precipitin reaction

Both in performing the test and in preparing the reagents, I was guided by the details of technique given by UHLENHUTH, NUTTALL and the school of MEZ, and followed scrupulously those which I learned personally from WASSERMANN. In my researches, I have adopted the method of adding a constant amount of immune serum to a series of progressively diminishing dilutions of precipitinogen.

A series of 12 test tubes (7 cm in length and 1 cm in diameter) is laid in a row in a rack. The first 10 tubes are marked progressively from to 25,600, meaning dilutions from 1 in 50 to 1 in 25,600. The eleventh is marked "E" for extract; the twelfth, which contains a solution of 0.9 percent sodium chloride, is marked NaCl. A second series of tubes, marked exactly like the first, is put in the second row of the rack. This second series represents the control. In tubes No. 2 to No. 11 of both series 0.5 cc of absolutely clear filtered and sterile 0.9 percent NaCl solution is put. With a sterile 1 cc pipette, 0.5 cc of very carefully neutralized and absolutely clear extract is put in tubes No. 1 and No. 11. Another 0.5 cc of the same extract is put in tube No. 2 of the front row of tubes. From tube No. 2, which now contains 1 cc of fluid, after well mixing the extract with the NaCl by sucking and expelling the fluid with the pipette, 0.5 cc is aspirated and transferred to tube No. 3. The same procedure is repeated up to tube No. 10, from which the extra 0.5 cc of the fluid is discarded. Thus,  each tube in the first row now contains 0.5 cc of fluid as specified above. In the second row of tubes, the same dilutions of precipitinogen are are repeated in exactly the same way as in the first. To each tube of the first row, with the exception of No. 11, is added 0.1 cc of the immune serum prepared against the homologous extract, by letting the fluid run Very slowly along the wall of the test tube. To each tube of the second row 0.1 cc of fresh normal rabbit serum is added in the same way. This control shows at once whether the precipitation obtained is a specific one or whether it is due to other causes, principally the acidity of the extract. Control No. 11 of the first and second rows will show that the extract by itself does not give rise to precipitation and control No. 12 of the first row will show that normal rabbit serum plus NaCl is also absolutely free of any precipitate. With a large number of controls, the security of the results is warranted. The test tubes are now put into an incubator at 37¡C. Before the reading of the reaction, in order to avoid autosuggestion on the part of the examiner, the identification marks on the test tubes are carefully concealed. A reading is taken every hour for the first 6 hours and then after 12 hours. As a rule, when the immune serum has a high titer, soon after the addition of the immune serum, a. slight turbidity is noticeable. This gradually increases and, after a few minutes, gives rise to fine precipitates which are very manifest already after 15 minutes. Naturally, if the reaction is a negative one, precipitins are not formed and the fluid in the test tubes remains absolutely clear even after 24 hours. At times, slight turbidities may be noticed, which are to be interpreted as negative or false reactions, since a definite precipitin reaction is characterized by the presence of fine, minute, whitish particles, like floccules, suspended in an absolutely clear fluid. These can be seen best through a magnifying lens by holding the test tube against a black ground and comparing it with the control tubes.

Conglutination reaction

The school of MEZ has adopted a modification to the precipitin reaction, which is called "conglutination." If to the dilution of precipitinogen and its immune serum, as used with the precipitin test, a small and constant amount of ox serum be added, owing to the property of the "conglutinins" contained in the fresh active ox serum, the precipitation of the extract with its immune serum is more evident and the time of the reading of this reaction is shortened from 24 to 2.5 hours. The following is the technique used by them: One cubic centimeter of the extract of a uniform concentration is put into each of 5 small test tubes. To the first tube is added 0.08 cc of the immune serum; to the second, 0.02 cc; to the third, 0.01 cc; and to the fourth, 0.005 cc. The tubes are then incubated for 2 hours at 37¡C and after this incubation each tube receives 0.4 cc of fresh ox serum. The tubes are again placed in the incubator and a reading taken every 20 minutes for 150 minutes.

Although MEZ and GOHLKE admit that the conglutination is "more, complicated than the precipitation," yet they do not think that the addition of ox serum in the conglutination reaction has ever been the cause of any disturbance in the results. In fact, they have carried out all the sero-diagnostic researches in plants with the conglutination method as the principal reaction, using the precipitin reaction as a control to the first.

As the precipitin phenomenon represents a specific reaction by itself, I endeavored to adhere to the standardized rules of technique of this reaction. The principle adopted in my research, of working with antigen containing large and determined amounts of albumin, has always given me very active immune sera. Although I adopted the precipitin reaction as the main means of investigation in my research, on several occasions I tried the conglutination method as advocated by the school of MEZ. In fact, I used the sera of both oxen and sheep and, although I must avow that a certain improvement in the reaction is noticeable in the way of reading, yet, I do not agree, after all, that the slight improvement in the reaction can justify the introduction of another factor, like an extra animal serum with all its unknown sources of possible errors, which might leave the validity of the results obtained open to criticism.

RUEHLE (1924), in a publication regarding the relationship of the members of the subfamily Prunoideae, made an anatomical study of the seeds, concluding that the development in the position of the seeds (Samenlagen)  of the Prunus domestica and the Prunus armeniaca is the same and that the relationship between the Pomaceae and the Prunoideae is is confirmed by the presence of the obturator, the nucellar and the chalazal haustoria.

Although in my researches, with the precipitin reaction, it was easy to obtain a positive result between the various members of the Prunus subfamily, I did not succeed in obtaining a similar result between the Prunoideae and the Pomaceae. The absence of a precipitin reaction between these two subfamilies seems to coincide with the difficulty encountered in the intergrafting of these two subdivisions of the same family. The explanation of this mysterious phenomenon among plants, which, by their anatomical and other morphological characteristics would appear to have had a common origin and development, but whose common origin is not confirmed by immunological reaction, is probably to be sought in the occurrence of stereoisomeric forms of proteins and carbohydrates, as REICHERT (1916) propounds in the following quotation:

"É.each kind of substance may exist in a number of forms, all of which forms have the same molecular formula and the same fundamental properties in common, but each in accordance with variations in intramolecular configuration has certain individualities which distinguish it from others. There are many known substances that exist in stereoisomeric forms and it has been found that the number of possible forms of each substance is dependent upon the possible number of variations of the arrangements of the molecular components in the three dimensions of space, or, in other words, of variations of molecular configuration, the possible number in case of each substance being capable of mathematical determination. Thus, we find that serum-albumin may exist in as many as a thousand million forms. Haemoglobin, the red coloring matter of vertebrate blood, is a far more complex carbon compound than serum-albumin, and theoretically may exist in forms whose number is beyond human conception, running into millions of millions. The same is true of starch."

In order to investigate more deeply this very important subject, a large series of grafting experiments ought to be carried out with plants giving the same precipitin reaction, and especially to try to solve the problem why it is more feasible to graft the quince than the apple on the pear, researches which unfortunately I did not have an opportunity of pursuing. With a methodical serological aid I think a more definite knowledge of this curious phenomenon could be obtained. As ZIEGENSPECK (1925) says:

"The serological system throws light over much which up to the present time was problematic and gives us good prospects for further investigation into the history of our vegetation."

RESULTS AND CONCLUSIONS

The results of my studies are tabulated in table 1 from which the following conclusions seem to be justified:

(1) Plants of the genus Citrus, known to be easily intergrafted,  gave a uniform positive precipitin reaction with an immune serum obtained from one of them.

(2) The subfamily of the Prunoideae, also known to intergraft successfully among themselves, although giving a uniform positive precipitin reaction among members of the same subfamily, gave a negative result when made to react with members of the subfamily Pomoideae. The Pomoideae seem to be an independent entity from the Prunoideae.

(3) The specimens of the Solanaceae examined, also known to intergraft, gave also a positive precipitin reaction.

(4) The control specimens of Beta vulgaris (beet), of Rheum officinale (rhubarb) and of Olea europaea (olive), belonging to three different families, gave always a negative result with each of the immune sera of the other families investigated in this research.

TABLE I

Precipitin reaction

PRECIPITINOGEN
OF THE

IMMUNE SERUM
AGAINST
C. decumana
TITER 1: 12,800

IMMUNE SERUM
AGAINST
P. amygdalus
TITER 1:12,800

IMMUNE SERUM
AGAINST
Pyrus malus
TITER 1: 6400

IMMUNE SERUM
AGAINST
Lycopersicum esculentum
TITER 1: 6400

NORMAL
RABBIT
SERUM

Family: Rutaceae

Genus: Citrus

C. decumana

+++

-

-

-

-

C. aurantium

+++

-

-

-

-

C. limonum

+++

-

-

-

-

C. vulgaris

+++

-

-

-

-

C. nobilis

+++

-

-

-

-

Family: Rosaceae

Subfamily: Prunoideae

P. amygdalus

-

+++

-

-

-

P. armeniaca

-

+++

-

-

-

P. persica

-

+++

-

-

-

P. domestica

-

+++

-

-

-

Family: Rosaceae

Subfamily: Pomoideae

Pyrus malus

-

-

+++

-

Pyrus communis

-

-

+++

-

-

Family: Solanaceae

Lycopersicum esculentum

-

-

-

+++

-

Solanum luberosum

-

-

-

+++

-

Family: Chenopodiaceae

Beta vulgaris

-

-

-

-

-

Family: Polygonaceae

Rheum officinale

-

-

-

-

-

Family: Oleaceae

Olea europaea

-

-

-

-

-

Note: + + + indicates a positive reaction.  
  - indicates a negative reaction.

LITERATURE CITED