LACTOSE IS DIGESTED BY LACTASE in the intestine, a single epithelial cell of which is enlarged 37,500 diameters in this scan·

ning electron micrograph made by Jeanne M. Riddle of the Wayne


State University School of Medicine. The cell, on the surface of

one of the villi that stud the lining of the intestine, is

in turn covered by innumerable line processes called microvilli.



Lactose is milk sugar; the enzyme lactase breaks it down. For want

of lactase most adults cannot digest milk. In populations that drink

milk the adults have more lactase, perhaps through natural selection

M ilk is the universal food of new­born mammals, but some hu­man infants cannot digest it be­ cause they lack sufficient quantities of lactase, the enzyme that breaks down lactose, or milk sugar. Adults of all ani­ mal species other than man also lack the enzyme-and so, it is now clear, do most human beings after between two and four years of age. That this general adult deficiency in lactase has come as a surprise to physiologists and nutrition­ ists can perhaps be attributed to a kind of ethnic chauvinism, since the few hu­ man populations in which tolerance of lactose has been found to exceed in­ tolerance include most northern Euro­ pean and white American ethnic groups.

Milk is a nearly complete human food, and in powdered form it can be con­ veniently stored and shipped long dis­ tances. Hence it is a popular source of protein and other nutrients in many programs of aid to nutritionally impov­ erished children, including American blacks. The discovery that many of these children are physiologically intolerant to lactose is therefore a matter of concern and its implications are currently being examined by such agencies as the U.S. Office of Child Development and the Protein Advisory Group of the United Nations System.

Lactose is one of the three major solid components of milk and its only carbo­ hydrate; the other components are fats and proteins. Lactose is a disaccharide composed of the monosaccharides glu­ cose and galactose. It is synthesized only by the cells of the lactating mammary gland, through the reaction of glucose with the compound uridine diphosphate galactose [see illustrations on next page]. One of the proteins found in milk, alpha-lactalbumin, is required for the synthesis of lactose. This protein appar­ ently does not actually enter into the

by Norman Kretchmer

reaction; what it does is "specify" the action of the enzyme galactosyl trans­ ferase, modifying the enzyme so that in the presence of alpha-lactalbumin and glucose it catalyzes the synthesis of lactose.

In the nonlactating mammary gland, where alpha-lactalbumin is not present, the enzyme synthesizes instead of lac­ tose a more complicated carbohydrate, N-acetyl lactosamine. Test-tube studies have shown that alpha-lactalbumin is manufactured only in the presence of certain hormones: insulin, cortisone, es­ trogen and prolactin; its synthesis is in­ hibited by the hormone progesterone. It is when progesterone levels decrease late in pregnancy that the manufacture of alpha-lactalbumin, and thus of lac­ tose, is initiated [see "Milk," by Stuart Patton; SCIENTIFIC AMERICAN, July, 1969].

The concentration of lactose in milk from different sources varies consider­ ably. Human milk is the sweetest, with 7.5 grams of lactose per 100 milliliters of milk. Cow's milk has 4.5 grams per 100 milliliters. The only mammals that do not have any lactose-or any other carbohydrate-in their milk are certain of the Pinnipedia: the seals, sea lions and walruses of the Pacific basin. If these animals are given lactose in any form, they become sick. (In 1933 there was a report of a baby walrus that was fed cow's milk while being shipped from Alaska to California. The animal suffered from severe diarrhea through­ out the voyage and was very sick by the time it arrived in San Diego.) Of these pinnipeds the California sea lion has been the most intensively studied. No alpha-lactalbumin is synthesized by its mammary gland. When alpha-lactalbu­ min from either rat's milk or cow's milk is added to a preparation of sea lion mammary gland in a test tube, however,

the glandular tissue does manufacture lactose.

In general, low concentrations of lac­ tose are associated with high concentra­ tions of milk fat (which is particularly useful to marine mammals). The Pa­ cific pinnipeds have more than 35 grams of fat per 100 milliliters of milk, com­ pared with less than four grams in the cow. In the whale and the bear (an an­ cient ancestor of which may also be an ancestor of the Pacific pinnipeds) the lactose in milk is low and the fat content is high.

�ctase, the enzyme that breaks down lactose ingested in milk or a milk

product, is a specific intestinal beta­ galactosidase that acts only on lactose, primarily in the jejunum, the second of the small intestine's three main seg­ ments. The functional units of the wall of the small intestine are the villus (composed of metabolically active, dif­ ferentiated, nondividing cells) and the crypt (a set of dividing cells from which those of the villus are derived). Lactase is not present in the dividing cells. It appears in the differentiated cells, spe­ cifically within the brush border of the cells at the surface of the villus [see il­ lustrations on page 74]. Lactase splits the disaccharide lactose into its two component monosaccharides, glucose and galactose. Some of the released glu­ cose can be utilized directly by the cells of the villus; the remainder, along with the galactose, enters the bloodstream, and both sugars are metabolized by the liver. Neither Gary Gray of the Stan­ ford University School of Medicine nor other investigators have been able to distinguish any qualitative biochemical or physical difference among the lac­ tases isolated from the intestine of in­ fants, tolerant adults and intolerant adults. The difference appears to be



LACTOSE, a disaccharide composed of the monosaccharides glucose and galactose, is the

carbohydrate of milk, the other major components of which are fats, proteins and water.

merely quantitative; there is simply very little lactase in the intestine of a lactose­ intolerant person. In the intestine of Pa­ cific pinnipeds, Philip Sunshine of the Stanford School of Medicine found, there is no lactase at all, even in infancy.

Lactase is not present in the intestine of the embryo or the fetus until the mid­ dle of the last stage of gestation. Its


activity attains a maximum immediate­ ly after birth. Thereafter it decreases, reaching a low level, for example, im­ mediately after weaning in the rat and after one and a half to three years in most children. The exact mechanism in­ volved in the appearance and disappear­ ance of the lactase is not known, but such a pattern of waxing and waning

I UDP-GALACTOSEI + ell G�L;!;U[g C:gO§:S�EII----��--=�[1 h!LA�C�T�O�S§EI + I UDP I

I UDP-GALACTOSEj + r-----"---� LA�t'8§r�fNE + IUDPI ===..!:.!!:�

1r DI LA�C�TO�S�E�I------------��------�»�I�G�LU�C�O�s�E I + �ACTOSE I SYNTHESIS OF LACTOSE in the mammary gland begins late in pregnancy when specific hormones and the protein alpha.lactalbumin are present. The latter modifies the enzyme galactosyl transferase, "specifying" it so that it catalyzes the synthesis of lactose from glu· cose and galactose (top). In the nonlactating gland the glucose takes part in a different reo action (middle). In intestine lactase breaks down lactose to glucose and galactose (bottom).


activity is common in the course of de­ velopment; in general terms, one can say that it results from differential ac­ tion of the gene or genes concerned.

Soon after the turn of the century the distinguished American pediatrician Abraham Jacobi pointed out that diar­ rhea in babies could be associated with the ingestion of carbohydrates. In 1921 another pediatrician, John Howland, said that "there is with many patients an abnormal response on the part of the intestinal tract to carbohydrates, which expresses itself in the form of diarrhea and excessive fermentation." He sug­ gested as the cause a defiCiency in the hydrolysis, or enzymatic breakdown, of lactose.

The physiology is now well estab­ lished. If the amount of lactose present­ ed to the intestinal cells exceeds the hy­ drolytic capacity of the available lactase (whether because the lactase level is low or because an unusually large amount of lactose is ingested), a portion of the lac­ tose remains undigested. Some of it passes into the blood and is eventually excreted in the urine. The remainder moves on into the large intestine, where two processes ensue. One is physical: the lactose molecules increase the particle content of the intestinal fluid compared with the fluid in cells outside the intes­ tine and therefore by osmotic action draw water out of the tissues into the in­ testine. The other is biochemical: the glucose is fermented by the bacteria in the colon. Organic acids and carbon di­ oxide are generated and the symptoms can be those of any fermentative diar­ rhea, including a bloated feeling, flatu­ lence, belching, cramps and a watery, explosive diarrhea.

At the end of the 1950's Paolo Du­ rand of the University of Genoa and Aaron Holzel and his colleagues at the UniverSity of Manchester reported de­ tailed studies of infants who were un­ able to digest lactose and who reacted to milk sugar with severe diarrhea, mal­ nutrition and even death. This work stimulated a revival of interest in lac­ tose and lactase, and there followed a period of active investigation of lactose intolerance. Many cases were reported, including some in which lactase inac­ tivity could be demonstrated in tissue taken from the patient's intestine by biopsy. It became clear that intolerance in infants could be a congenital condi­ tion (as in Holzel's two patients, who were Siblings) or, more frequently, could be secondary to various diseases and other stresses: cystic fibrosis, celiac disease, malnutrition, the ingestion of certain drugs, surgery and even non-


specific diarrhea. During this period of investigation, it should be noted, in­ tolerance to lactose was generally as­ sumed to be the unusual condition and the condition worthy of study.

In 1965 Pedro Cuatrecasas and his colleagues and Theodore M. Bayless and Norton S. Rosensweig, all of whom were then at the Johns Hopkins School of Medicine, administered lactose to Amer­ ican blacks and whites, none of whom had had gastrointestinal complaints, and reported some startling findings. Where­ as only from 6 to 15 percent of the whites showed clinical symptoms of in­ tolerance, about 70 percent of the blacks were intolerant. This immediately sug­ gested that many human adults might be unable to digest lactose and, more specifically, that there might be signifi­ cant differences among ethnic groups. The possibility was soon confirmed: C. C. Cook and S. Kajubi of Makerere University College examined two differ­ ent tribes in Uganda. They found that only 20 percent of the adults of the cat­ tle-herding Tussi tribe were intolerant to lactose but that 80 percent of the non­ pastoral Canda were intolerant. Soon one paper after another reported a gen­ eral intolerance to lactose among many ethnic groups, including Japanese, other Orientals, Jews in Israel, Eskimos and South American Indians.

In these studies various measures of intolerance were applied. One was the appearance of clinical symptoms-flatu­ lence and diarrhea-after the ingestion of a dose of lactose, which was generally standardized at two grams of lactose per kilogram (2.2 pounds) of body weight, up to a maximum of either 50 or 100 grams. Another measure was a finding of low lactase activity (less than two units per gram of wet weight of tissue) deter­ mined through an intestinal biopsy after ingestion of the same dose of lactose. A third was an elevation of blood glu­ cose of less than 20 milligrams per 100 milliliters of blood after ingestion of the lactose. Since clinical symptoms are variable and the biopsy method is in­ convenient for the subject being tested, the blood glucose method is preferable. It is a direct measure of lactose break­ down, and false-negative results are rare if the glucose is measured 15 minutes after lactose is administered.

By 1970 enough data had been ac­ cumulated to indicate that many more groups all over the world are intolerant to lactose than are tolerant. As a matter of fact, real adult tolerance to lactose has so far been observed only in north­ ern Europeans, approximately 90 per­ cent of whom tolerate lactose, and in the











3 4






i I , I



6 7

CONCENTRATION OF LACTOSE varies with the source of the milk. In general the less

lactose, the more fat, which can also be utilized by the newborn animal as an energy source.

5 1(\ 4

� 3 :> i= o « w (f)

� o 2 :5

o 10 o




\ \

� -- _. 20 30 40

DAYS AFTER BIRTH ADULT LACT ASE is present in mammals other than man, and in most humans, in the fetus before

birth and in infancy. The general shape of the curve of enzyme activity, shown here for the

rat, is about the same in all species. Enzyme activity, given here in relative units, is de·

termined by measuring glucose release from intestinal tissue in the presence of lactose.



VILu -========mn&:L







LYMPH VESSEL----_��dl.Ja


DIFFERENTIATII\IG1-"---__ .:g,....,.--Ui CELLS

LAMINA PROPRIA r:::;;?:5;;:;��"" CRYPT-------F��----��--����"·. DIVIDING CELLS

I7 --'--____ --.:--.:_�

WALL OF SMALL INTESTINE, seen in longitudinal section (top), has outer muscle lay. ers, a submucosa layer and an inner mucous membrane. The mucous membrane (bottom) has a connective·tissue layer (lamina propria), which contains blood and lymph capillaries,

and an inner surface of epithelial cells. The cells multiply and differentiate in the crypts

and migrate to the villi. At what stage the lactase is manufactured is not known; it is found

primarily in the microvilli, which constitute the brush border of the differentiated cells.


membel"s of two nomadic pastoral tribes in Africa, of whom about 80 percent are tolerant. Although many other gen­ erally tolerant groups will be found, they will always belong to a minority of the human species. In this situation it is clearly more interesting and potentially more fruitful to focus the investigation on tolerant people in an effort to explain adult tolerance, a characteristic in which man differs from all other mammals.

There are two kinds of explanation of adult tolerance to lactose. The first, and perhaps the most immediately apparent, originates with the fact that most people who tolerate lactose have a history of drinking milk. Maybe the mere pres­ ence of milk in the diet suffices to stimu­ late lactase activity in the individual, perhaps by "turning on" genes that en­ code the synthesis of the enzyme. In­ dividual enzymatic adaptation to an en­ vironmental stimulus is well known, but it is not transferable genetically. The other explanation of tolerance is based on the concept of evolution through natural selection. If in particular popu­ lations it became biologically advan­ tageous to be able to digest milk, then the survival of individuals with a genetic mutation that led to higher intestinal lactase activity in adulthood would have been favored. An individual who de­ rived his ability to digest lactose from this classical form of Darwinian adapta­ tion would be expected to be able to transfer the trait genetically.

These two points of view have become the subject of considerable contro­

versy. I suspect that each of the expla­ nations is valid for some of the adult tolerance being observed, and I should like to examine both of them.

The possibility of individual adapta­ tion to lactose has been considered since the beginning of the century, usually through attempts to relate lactase ac­ tivity to the concentration of milk in the diet of animals. Almost without excep­ tion the studies showed that although there was a slight increase in lactase activity when a constant diet of milk or milk products was consumed, there was no significant change in the characteris­ tic curve reflecting the developmental rise and fall of enzymatic activity. Re­ cently there have been reports pointing toward adaptation, however. Some stud­ ies, with human subjects as well as rats, indicated that continued intensive feed­ ing of milk or lactose not only made it possible for the individual to tolerate the sugar but also resulted in a measurable increase in lactase activity. The discrep­ ancy among the findings could be partly


attributable to improvement in meth­ ods for assaying the enzyme activity.

On balance it would appear that in­ dividual adaptation may be able to ex­ plain at least some cases of adult toler­ ance. I shall cite two recent studies. John Codell, working in Lagos, selected six Nigerian medical students who were absolutely intolerant to lactose and who showed no physiological evidence of lac­ tose hydrolysis. He fed them increasing amounts of the sugar for six months. Codell found that although the students did develop tolerance for the lactose, there was nevertheless no evidence of an increase of glucose in the blood-and thus of enzymatic adaptation-following test doses of the sugar. The conjecture is that the diet brought about a change in the bactertlll flora in the intestine, and that the ingested lactose was being me­ tabolized by the new bacteria.

In our laboratory at the Stanford School of Medicine Emanuel Lebenthal and Sunshine found that in rats given lactose the usual pattern of a develop­ mental decrease in lactase activity is maintained but the activity level is some­ what higher at the end of the experi­ ment. The rise in activity does not ap­ pear to be the result of an actual increase in lactase synthesis, however. We treat­ ed the rats with actinomycin, which pre­ vents the synthesis of new protein from newly activated genes. The actinomycin had no effect on the slight increase in lactase activity, indicating that the mechanism leading to the increase was not gene activation. It appears, rather, that the presence of additional amounts of the enzyme's substrate, lactose, some­ how "protects" the lactase from degra­ dation. Such a process has been noted in many other enzyme-substrate systems. The additional lactase activity that re­ sults from this protection is sufficient to improve the rat's tolerance of lactose, but that additional activity is dependent on the continued presence of the lactose.

Testing the second hypothesis-that adult lactose tolerance is primarily the result of a long-telID process of ge­ netic selection-is more complicated. It involves data and reasoning from such disparate areas as history, anthropology, nutrition, genetics and sociology as well as biochemistry.

As I have noted, the work of Cuatre­ casas, of Bayless and Rosensweig and of Cook and Kajubi in the mid-1960's pointed to the likelihood of Significant differences in adult lactose tolerance among ethnic groups. It also suggested that one ought to study in particular black Americans and their ancestral pop-

ulations in Africa. The west coast of Af­ rica was the primary source of slaves for the New World. With the objective of studying lactose tolerance in Nigeria, we developed a joint project with a group from the University of Lagos Teaching Hospital headed by Olikoye Ransome-Kuti.

The four largest ethnic groups in Ni­ geria are the Yoruba in western Nigeria, the rbo in the east and the Fulani and Hausa in the north. These groups have different origins and primary occupa­ tions. The Yoruba and the Ibo differ somewhat anthropometrically, but both are Negro ethnic groups that probably came originally from the Congo Basin; they were hunters and gatherers who be­ came farmers. They eventually settled south of the Niger and Benue rivers in an area infested with the tsetse fly, so that they never acquired cattle (or any other beast of burden). Hence it was not until recent times that milk appeared in their diet beyond the age of weaning. After the colonization of their part of Nigeria by the British late in the 19th century, a number of Yoruba and lbo, motivated by their intense desire for education, migrated to England and northern Europe; they acquired West­ ern dietary habits and in some cases Western spouses, and many eventually returned to Nigeria.

The Fulani are Hamites who have been pastoral people for thousands of years, originally perhaps in western Asia and more recently in northwestern Afri­ ca. Wherever they went, they took their cattle with them, and many of the Fu­ lani are still nomads who herd their cat­ tle from one grazing ground to anoth­ er. About 300 years ago the Fulani ap­ peared in what is now Nigeria and waged war on the Hausa. (The Fulani also tried to invade Yorubaland but were defeated by the tsetse fly.) After the in­ vasion of the Hausa region some of the Fulani moved into villages and towns.

As a result of intermarriage between the Fulani and the Hausa there ap­ peared a new group known as the town­ Fulani or the Hausa-Fulani, whose mem­ bers no longer raise cattle and whose in­ gestion of lactose is quite different from that of the pastoral Fulani. The pastoral Fulani do their milking in the early morning and drink some fresh milk. The milk reaches the market in the villages and towns only in a fermented form, however, as a kind of yogurt called nono. As the nono stands in the morning sun it becomes a completely fermented, wa­ tery preparation, which is then thick­ ened with millet or some other cereal. The final product is almost completely





" U

WATER ---t-t--);.


r') (


. BACTERIAlr � �

(;><E ..... '-+-t-­ CARBON DIOXIDE�

I <:J

___ ---i-;--1> 0 ACID .... 0 ./ ./ <1 ......

\l 0


0 '- ,�

:E � z w o o � o


9 o u

DIGESTION OF LACTOSE is accomplished

primarily in the jejunum, where lactase

splits it into glucose and galactose. Some

glucose is utilized locally; the rest enters

the bloodstream with the galactose and both

are utilized in the liver. In the absence of

enough lactase some undigested lactose en·

ters the bloodstream; most goes on into the

ileum and the colon, where it draws water

from the tissues into the intestine by osmot·

ic action. The undigested lactose is also fer· mented by bacteria in the colon, giving rise

to various acids and carbon dioxide gas.



$150 a: W 1-125 :J

-'-' --- +- - - -- -- -_ . .. _._-- :J

w=hoo �::;; 00 ::Jo 75 ....J� <!la: OW 50 00. OCl)

0-�t- 1 � �-----

I I '""'\

/ � / I / I

-- --r---

- --.. -




ffi� 25 I

)----1 -- a: g 0 ....J 0 �

30 60 90 120 0 30 60 90 120 0 30 60 90 120

LACTOSE INTOLERANCE is determined by measuring blood glu·

cose after ingestion of lactose. The absence of a significant rise in

blood glucose after lactose ingestion (color) as contrasted with a

rise in blood glucose after ingestion of sucrose, another sugar

(black), indicates that a Y oruba male ( left) and an American Jewish male (middle) are lactose· intolerant. On the other hand,

the definite rise in blood glucose after ingestion of lactose in a

Fulani male (right) shows that tbe Fulani is tolerant to lactose.

free of lactose and can be ingested with­ out trouble even by a person who can­ not digest lactose.

We tested members of each of these Nigerian populations. Of all the Yorubas above the age of four who were tested, we found only one person in whom the blood glucose rose to more















than 20 milligrams per 100 milliliters following administration of the test dose of lactose. She was a nurse who had spent six years in the United Kingdom and had grown accustomed to a British diet that included milk. At first, she said, the milk disagreed with her, but later she could tolerate it with no ad­ verse side effects. None of the rbos who

were studied showed an elevation of glucose in blood greater than 20 milli­ grams per 100 milliliters. (The major problem in all these studies is deter­ mining ethnic purity. All the Yorubas and rbos who participated in this portion of the study indicated that there had been no intermarriages in their families.) Most of the Hausa and Hausa-Fulani


INTOLERANCE VARIES WIDELY among populations. The bars

are based on tests conducted by a number of investigators by dif­

ferent methods; they may not be strictly comparable or accurate·

ly reflect the situation in entire populations. Among the groups

studied to date lactose intolerance is prevalent except among

northern Europeans (and their descendants) and herders in Africa.



(70 to 80 percent) were intolerant to lactose. In contrast most of the nomadic Fulani (78 percent) were tolerant to it. In their ability to hydrolyze lactose they resem bled the pastoral Tussi of Uganda and northern Europeans more than they resembled their nearest neighbors.

Once the distribution of lactose in­ tolerance and tolerance was determined in the major Nigerian populations, we went on to study the genetics of the situation by determining the results of mixed marriages. One of the common marriages in western Nigeria is between a Yoruba male and a British or other northern European female; the reverse situation is less common. Our tests showed that when a tolerant northern European marries a lactose-intolerant Yoruba, the offspring are most likely to be lactose-tolerant. If a tolerant child resulting from such a marriage marries a pure Y Oluba, then the children are also predominantly tolerant. There is no sex linkage of the genes involved: in the few cases in which a Yoruba fe­ male had married a northern European male, the children were predominantly tolerant.

On the basis of these findings one can say that lactose tolerance is trans­ mitted genetically and is dominant, that is, genes for tolerance from one of the parents are sufficient to make the child tolerant. On the other hand, the children of two pur� Yorubas are always intoler­ ant to lactose, as are the children of a lactose-intolerant European female and a Yoruba male. In other words, intoler­ ance is also transmitted genetically and is probably a recessive trait, that is, both parents must be lactose-intolerant to produce an intolerant child. When the town-dwelling royal line of the Fulani was investigated, its members were all found to be unable to digest lactose­ except for the children of one wife, a pastoral Fulani, who were tolerant.

Among the children of Yoruba-Euro­ pean marriages the genetic cross oc­ curred one generation ago or at the most two generations. Among the Hausa-Fu­ lani it may have been as much as …