Frozen Evolution. Or, that’s not the way it is, Mr. Darwin. A Farewell to Selfish Gene.

When studying the effect of toxoplasmosis on the course of pregnancy we discovered another phenomenon, which we had not previously expected and whose existence shocked us. Šárka Kaňková, my undergraduate student at the time, noticed that latent toxoplasmosis also influences the sex ratio at birth (SRB); that is, the ratio of sons to daughters (92). Specifically, we found that Toxo positive mothers gave birth to more boys than girls (Fig. 45). But this relationship is only true for women with high or medium levels of antibodies, i.e. for women who were infected for a relatively short amount of time – an estimated 1-4 years. In contrast, women with very low antibody levels, i.e. those who were infected long ago, had predominantly daughters. The observed differences were highly significant; for example, in 111 women with the highest antibody levels (women who were probably infected 1-2 years before pregnancy), the ratio of sons to daughters was 260 to 100 (Fig. 46).

Fig. 45 The prevalence of sons in mothers infected by Toxoplasma. In all three obstetric and gynecological clinics, Toxo-infected mothers (gray bars) gave birth to significantly more sons than daughters. With the exception of the Gyncentrum, uninfected women (empty bars) also gave birth to slightly more sons. We hypothesize that this is because Gest clinics are private clinics where mothers pay for obstetrical care; most likely these women have a higher economic status. In such a case, we would expect a shift to a higher proportion of sons due to the Trivers-Willard effect. Numbers inside the bars represent the number of women in the group. The statistical significance of the effect toxoplasmosis on the gender index at birth was 0.001.

Fig. 46 The relationship between the gender index at birth (the fraction of sons in the offspring, or also the probability of having a son) and the concentration of antibodies against Toxoplasma – the latter correlates with the time elapsed after infection. The level of antibodies was determined using indirect immunofluorescence. Sixteen women with the lower concentration of antibodies – women infected longest ago – had more daughters than sons (a gender index lower than 0.5). In contrast, 132 women with the highest concentrations of antibodies (titer 64 and higher), women infected relatively recently, gave birth to more sons than daughters (gender index 0.68). This group of 132 includes 111 women with a titer of 128 and higher, who had a gender index of 0.72, a ratio of 2.6 daughters per son. Relationship of gender index on concentration of antibodies against Toxoplasma was highly statistically significant (P = 0.002).

Šárka Kaňková later confirmed this effect in two independent tests using mice as models (93). The first experiment had 60 total mice, and the second had 80; in each experiment, half the mice were infected and the other half uninfected. About six weeks after infection, we put the males and females together, and then looked at the sex ratio of each litter – that is, the ratio of male to female offspring of each infected or uninfected mouse. Again we found that 2-3 months after infection, the mice gave birth to significantly more sons than daughters. With time after infection the ratio of male to female offspring evened out and began to reverse, so that towards the end of the experiment, mice which had been infected 4-6 months gave birth to more daughters than sons – that it, even more daughters than the control mice. The results obtained on mice, therefore, precisely correspond with the trends observed in pregnant women. In the earlier phases of latent toxoplasmosis we noted a predominance of sons, and in the later phases a prevalence of daughters.

The decrease of the sex ratio in later phases of infection – i.e. the birth of more daughters than sons – might be due to cumulative negative effects of latent toxoplasmosis on the host’s state of health. The symptoms of such health impairment are discussed in chapter 10; and the Trivers-Willard effect (a decreased sex ratio in females with poor health) will be discussed at the end of this chapter. The increased sex ratio in the earlier phases of toxoplasmosis is much more interesting. It may stem from one of three things: a side-effect of Toxoplasma infection; a side-effect of Toxo’s manipulation of the immune system; or even a direct manipulation by the parasite of the sex ratio, in an effort to increase its chances of transmission to another host. It certainly isn’t easy to recognize, whether we’re observing a manipulation advantageous to Toxoplasma, or whether it’s only a side effect of infection by this parasite. In species in which a mother with latent toxoplasmosis cannot transmit the parasite to the fetus, it might be beneficial for Toxoplasma for mothers with acute toxoplasmosis (which can be transmitted to offspring) to give birth to sons. In most mammals, the males are the ones who migrate the greatest distances, and who therefore can act as a better vector for spreading Toxoplasma to new, not yet infected population. But in some species of mice, congenital transmission of toxoplasmosis from the mother to offspring can occur, and in this case it would be more advantageous for the parasite if the infected mother gave birth to more daughters, who could then pass down the parasite to their children Since it isn’t clear what host species Toxoplasma is originally adapted to (see chapter 10), it’s difficult to judge what is and isn’t beneficial for the parasite.

It’s not too clear how Toxoplasma affects the sex of newborns. What seems most likely to us is that Toxoplasma induces immunosuppression, i.e. it lowers the immunity of its host, and thereby actually saves part of the mother’s embryos, which would have been aborted. It’s known that when eggs are fertilized by sperm, the sex ratio in humans is about 1:1. But when the fertilized egg nests in the uterus, this ratio significantly changes in the favor of sons; specifically about 2.5 to 3 times more male than female embryos begin development. But in the other phases of pregnancy, male embryos are much more susceptible to being aborted (Box 86 Males as mutation gargage-cans and cheap testing material).

For this reason, when it comes to the birth, the ratio of males to females isn’t 2.5 or 3 to 1, in favor of the males, but only 1.06 to 1 in favor of the males. And in getting rid of male embryos, the immune system of the mother plays a significant role. This is because male embryos are more immunogenic (meaning that they set off a strong response of the mother’s immune system) than are female embryos. Besides the antigens found in both sons and daughters, there also exist H-Y antigens, which come from proteins typical to male tissue. If Toxoplasma can induce immunosuppression, then it lowers the effectiveness of aborting male fetuses, and many more males will complete their development.

The same mechanism may also occur in the case of another interesting phenomenon, described by Czech parasitologists in the 60s. Otto Jírovec, the founder of modern Czech parasitology, and his coworkers showed that children with Down’s syndrome often, about 84% of them, have Toxo positive mothers. In a control group, which gave birth to children without Down’s syndrome, the prevalence of toxoplasmosis was about 30%. In the fathers of children without Down’s syndrome the occurrence of toxoplasmosis was also only 30%. The dramatically greater prevalence of toxoplasmosis in mother’s of children with Down’s syndrome could again be explained by the fact that most fetuses with chromosomal defects, in this case, fetuses which have an extra copy of the 21^st chromosome, are aborted in healthy women, specifically aborted in early pregnancy. It’s known that in the first weeks of pregnancy there occurs some sort of checking of the embryo’s quality. Seeing as Toxoplasma suppresses immune activity, it may be able to lower the strictness of this control. As a result, a much greater percent of damaged fetuses survive, and that is why Toxo positive women give birth to more children with defects. From an ethical standpoint, it’s an interesting question whether the parents of children with Down’s syndrome should curse or thank Toxoplasma – the parasite doesn’t actually cause Down’s syndrome in the children, but only saves their lives.

We originally deduced this idea, that latent toxoplasmosis could be associated with immunosuppression, from the results of our studies regarding changes in the sex ratio at birth, as well as results described by Otta Jírovec. And recently we got the direct data to prove it. Working with Ilja Stříž of the Prague Institute for Clinical and Experimental Medicine (IKEM), we discovered that both men and women with latent toxoplasmosis have fewer B cells (white blood cells that produce antibodies) than do uninfected persons. In addition, we found that infected men have fewer white blood cells in general – and specifically fewer natural killer (NK) cells and monocytes. Interestingly, infected women have a higher number of white blood cells, including NK cells and monocytes. It’s possible that the differences between men and women are related to the fact that each gender has an opposite shift in testosterone associated with toxoplasmosis. Testosterone lowers immunity, and as we discovered, infected men experience an increase in the hormone, whereas infected women see a decrease. The fact that infected women have a higher number of certain cells involved in immune reaction does not necessarily mean that aren’t experiencing immunosuppression. Šárka Kaňková conducted experiments on the matter, collaborating with Vladimír Holáň and Alena Zajícová from the Institute of Molecular Genetics. These experiments showed that female mice infected with Toxoplasma have higher levels of certain regulatory molecules which stimulate immune reaction, such as interleukin 12. However, their macrophages (white blood cells, the purpose of which is to devour large foreign objects or dead parts of the body’s own tissue), and lymphocytes (white blood cells which produce antibodies (B cells), or which kill foreign cells (T cells)) do not react appropriately to these stimuli. In contrast with the macrophages of uninfected mice, these macrophages produced less nitric oxide (a substance which the macrophage needs to kill the intruder). In addition, the spleen cells (which include a variety of white blood cells) produced less interleukin 2 (a substance that activates lymphocytes, namely T cells) and divide less than in uninfected mice when exposed to foreign cells from another strain of mice. The results were unambiguous – while the immune system of mice infected with Toxoplasma could recognize foreign antigens, it was not able to react to them (86). It’s particularly clear that the cellular mediated immunity is damaged, and this is precisely what part of immune system we suspect could be responsible in uninfected mice and women for aborting defective male embryos. Now I will dare to paraphrase the famous last sentence of the article in which Watson and Crick explained the structure of DNA: “It did not escape our attention that the existence of immunosuppression in a significant part of the population” (in people with latent toxoplasmosis) “could also have serious consequences for health.” (Didn’t I warn you in the first chapter, that I’m modesty itself?)

Now that we understand the probable mechanism behind the effect, we can return to the question of whether the shift in the offspring’s sex ratio at birth is due to Toxoplasma’s manipulatory activity, or merely a side effect of the disease. The second possibility is starting to seem more likely. Immunosuppression, particularly the suppression of the cell immunity that targets intracellular parasites, certainly helps Toxoplasma survive in the host organism, and the effect of lowering cell immunity on the sex ratio at birth could just be a side-effect of this immunosuppression. But nature doesn’t always abide by reason nor the popular principle of Occam’s razor, so who knows if this is really true (Box 87 Occam’s razor).

When evaluating the data of our studies focused on the change in the SRB of Toxo positive mothers, we stumbled upon another interesting phenomenon, though it has nothing to do with Toxoplasma (yes, such effects do exist). It’s known that in many animals the gender of the newborn depends on the health and social status of the mother. Females who are healthy and dominant in the social hierarchy have more sons, whether females with a low social standing and worse state of healthy have more daughters. The so-called Trivers–Willard hypothesis explains that there is much greater competition to reproduce among males than among females. In many species, only the top males reproduce, leaving behind an enormous number of offspring. The weak males don’t mate at all. In females, this relationship between fitness and number of offspring isn’t as crucial, since even weaker females usually reproduce. But if a female is healthy and in high social standing, she can invest enough resources in her offspring, so they will be more fit. In sons, fitness is important, for it can decisively influence their reproductive potential – again, only the top males will reproduce. So it pays off for females in high social standing to give birth to mostly sons, and for less fit females to invest in mostly daughters.

The Trivers-Willard hypothesis was tested several times on humans. It was discovered that even this atypical species likely fits the hypothesis, but unlike the studies on other animals, the results aren’t entirely convincing. There are only a couple of studies which show that woman with higher social standing really give birth to more sons than daughters. One such study was conducted on European nobility, which has detailed family trees. It was discovered that, unlike in the normal population, sons predominate. Similarly, it was discovered that American presidents, as well as high-ranking generals and multimillionaires, have a greater ratio of sons. Unfortunately, only a few studies have been conducted on humans. We were pleased to find that even our results indirectly confirm the applicability of this hypothesis to humans (92). It turned out that in our sample of women, the highest percentage of sons was among women who went to the most expensive local obstetrics and gynecology clinic, followed by women of the cheaper local clinic, and then those of the cheapest. But even the women who went to the cheapest local clinic gave birth to more sons than did women anywhere in Prague. Most of the Prague women were under the care of institutes originally run by the state, where they don’t have to pay for basic health care. This correlation between the expensiveness of obstetrical care and the sex ratio at birth was also statistically significant, and in our opinion represents another independent confirmation of the applicability of the Trivers-Willard hypothesis to humans. One may expect that women who go to a pricier local clinic generally come from circles that are socially and economically better-off than those frequented by women who attend a free clinic.

In chapter XVI, we saw that some effects of latent toxoplasmosis occur only in Rh negative people, and that humans of the Rh+ blood group are temporarily (Rh positive homozygotes) or even permanently (Rh positive heterozygotes) protected against them. Of course we were interested to know whether the protective effect in Rh positives also applies in the case of Toxo’s effect on pregnancy. When we studied the effect of toxoplasmosis on the sex ratio at birth, we found no differences between Rh positives and negatives. On the other hand, we saw substantial differences in Toxo’s effect on weight gain during pregnancy (Fig. 47). Put simply, out of four possible combinations – Rh negative Toxo negatives, Rh negative Toxo positives, Rh positive Toxo negatives, and Rh positive Toxo positives – only one is significantly different from the others. It is the group of Rh negative women with latent toxoplasmosis (94). In the first trimester, these women have almost twice the weight gain as do the other group of women. Around the 16^th week of pregnancy, they have about a 1.6 kilograms greater weight gain, which is a statistically significant difference that remains constant until the end of pregnancy. Yet the length and weight of their newborns are not significantly different, which shows that the greater weight gain of the Rh negative Toxo positives is due to the greater weight gain of the mother – likely in her uterus, amniotic sac or amniotic fluid – not a higher weight of the embryo itself.

Fig. 47 The abnormal weight gain of Rh negative Toxoplasma positive women during pregnancy. At the 16^th week of pregnancy, the weight gain in this group was almost twice as big as that in other women. This is the strongest effect of latent toxoplasmosis on the human organism that we’ve observed in our 20 years of study. Rh negative women are indicated by black circles, and Rh positive women by unshaded ones. The study examined 646 Rh positive Toxo negatives, 138 Rh negative Toxo negatives, 167 Rh positive Toxo positives and 27 Rh negative Toxo positives. With the exception of the 36^th week, the effect was always highly statistically significant.

It’s known that greater weight gain in early pregnancy usually signifies a defect in fetal development. We see it, for example, in mothers who smoke. Therefore, our results again indicate that Rh positivity (though the mechanism is still unknown) protects the mothers against negative effects of latent toxoplasmosis, this time against the bad effects on pregnancy. In Rh negative women, the effect of toxoplasmosis on pregnancy is unusually strong. The weight gain of Rh negative Toxo positive women in the 16^th week is about twice that in other women – the strongest effect of toxoplasmosis we have seen in all our years of study. It will certainly be essential to determine whether Rh positivity protects only against the negative effects of toxoplasmosis, or whether its influence extends to other factors, such as smoking, diabetes, or obesity.