Effect of Spirulina maxima and Its Protein Extract on Micronuclei Induction by Hydroxyurea in Pregnant Mice and Their Fetuses

Abstract

The purpose of the present report was to determine the inhibitory effect of Spirulina maxima (Sm) and its protein extract (PE), mainly consisting of C-phycocyanin, on the increase in micronuclei and bone marrow cytotoxicity induced by hydroxyurea (HU) in pregnant mice and their fetuses. The two tested antimutagenic agents were administered daily from day 10 to day 18 of pregnancy, and HU (300 mg/kg) was administered once on day 16 of the assay. The experimental design also included mice that were administered only Sm or PE (1000 and 400 mg/kg, respectively), two control groups that were administered with vehicles (water and 0.5% Tween 80), and one additional group that was treated solely with HU. Blood samples from the pregnant mice and their fetuses were examined at day 19 of pregnancy. Significant increases in the number of micronucleated polychromatic erythrocytes and in the total number of micronucleated erythrocytes were observed in all HU-treated animals. In contrast, similarly low numbers of micronuclei were observed in the two control groups and in the groups treated with Sm and PE alone. The administration of Sm (100, 500, and 1000 mg/kg) and PE (100, 200, and 400 mg/kg) to HU-treated animals conferred moderate genotoxic protection (∼30%) and some protection against the cytotoxicity induced by HU in mice. The obtained results provide new information regarding the capacity of the tested agents to confer protection to adult mice and transplacentally, as well as on a specific subclass of micronuclei.

Introduction

S pirulina maxima (Sm) is a filamentous unicellular blue-green alga belonging to the Oscillatoriaceae family that usually grows in alkaline waters in Africa, Asia, and America. The plant has been largely used as food due to its high protein, vitamin, mineral, and essential fatty acid content.^1,2 Moreover, the alga has been reported to possess a number of biomedical properties that are potentially beneficial for human health, such as antiviral attributes, preventive capacity against the development of anemia and fatty liver disease, and antihyperlipidemic as well as antihypertensive properties.³ It is also interesting to note that Sm has in vitro and in vivo antioxidant effects as well as antigenotoxic and anticarcinogenic properties.^4
–6

Protein extracts (PE) of the algae, which are mainly composed of C-phycocyanin, allophycocyanin, and phycoerythrin, have also been reported to possess anti-inflammatory, antioxidant, and neuroprotective activities, suggesting their potential application in biotechnology, food, and medicine.^7,8 In fact, the antioxidant effect of PE was found to be as strong as that determined for phycocyanin.⁹

In a previous report, we demonstrated that Sm and its PE were able to inhibit the congenital malformations induced in mice by hydroxyurea (HU), a chemical known for its oxidative properties, teratogenic capacity, and inhibition of embryonic DNA synthesis. Moreover, we also showed that the mechanism of action involved in such inhibition was most likely related to the antioxidant potential of the tested agents.¹⁰

Based on the information described above, we aimed to explore the participation of these two potentially protective agents in other biological activities. Therefore, in the present report, we evaluated the capacity of Sm and its PE to inhibit the genotoxic damage induced by HU in both pregnant mice and their fetuses. For this purpose, we studied the effect of these agents on the induction of micronuclei, which are cytological abnormalities that represent broken chromosomes or whole chromosomes that have abnormally segregated during cellular division. Micronuclei are reliable biomarkers for genomic instability and genotoxic exposure that are usually included in most batteries to determine genetic damage or protection.¹¹

A few studies in mice have demonstrated the protective effect of Spirulina against the induction of micronuclei by gamma radiation and chemical mutagens;^12,13 however, no reports have been published regarding its effect on HU-induced damage or its transplacental protective potential. Additionally, no information regarding the effect of Spirulina PE on the genotoxicity of HU has been published.

Materials And Methods

Chemicals, animals, and protein extraction procedure

Methanol, phosphate-buffered saline (PBS), Tween 80, and HU were obtained from J.T. Baker (Mexico City, Mexico). Giemsa stain was purchased from Merck S.A. (Mexico City, Mexico). Sm was supplied by Alimentos Esenciales para la Humanidad, S.A. de C.V. (Mexico City, Mexico). The algae corresponded to the bulk production batch SDW-9714, the quality of which is standardized according to the nutritional information provided by the manufacturer. For the present assay, the Sm powder was diluted in 0.5% Tween 80 (in water) immediately before use.

The genotoxic/antigenotoxic assay was carried out in CD-1 male mice weighing 30 g and female mice weighing 26 g. The animals were maintained at a constant temperature (22±2°C) and humidity (50±10%) under artificial illumination between 8:00 a.m. and 8:00 p.m. The mice were fed with pellets (Rodent Laboratory Chow 5001, Purina, Missouri, USA) and water ad libitum. The experimental protocol was approved by the Committee of Ethics and Security of the National School of Biological Sciences.

The present study corresponds to the second part of a previously published report;¹⁰ therefore, the experimental conditions, including the doses of both agents and the procedure for obtaining PE from Sm, were performed as previously described. Briefly, 1 g of Sm was suspended in 10 mL of PBS and stored for 12 h in the dark; subsequently, the suspension was centrifuged twice at 13,000 rpm (15 min each time, Sorvall RC-5B; Du Pont Instruments, Newtown, CT, USA) and twice at 35,000 rpm (30 min each time; L-80 Beckman Coulter, Brea, CA, USA). After each centrifugation step, the green precipitate was removed and the blue supernatant was centrifuged again. The obtained PE was dialyzed in the dark (at 4°C) against a 0.01 M sodium phosphate solution for 5 days using a 12-kDa membrane before lyophilization. The presence of C-phycocyanin in the extract was confirmed by a sodium dodecyl sulfate–polyacrylamide gel electrophoresis procedure that included a molecular weight marker of the specific protein. Moreover, we observed that 620 nm was the wavelength of maximum absorption for the PE extracts and that the ratio of A₆₂₀/A₂₈₀ was equal to 2. These two characteristics also identified the presence of C-phycocyanin.¹⁴

Experimental design

Before performing the measurements, the animals were divided into groups of five pregnant mice and groups of six fetuses per mother, following the recommendations of the Organization for Economic Co-operation and Development (414) for studies of prenatal developmental toxicity; some modifications were introduced as described by Warner et al. ¹⁵ For the assay, we also employed the methodology previously utilized by our laboratory.¹⁶ Mouse males were placed together with females for 2 h before initiating the illumination period. Dams that presented vaginal plugs were considered as being at day 1 of pregnancy; subsequently, these dams were randomly distributed for the experimental treatment. The vehicle for HU and PE was water, whereas Sm was diluted in 0.5% Tween 80 in water. Thus, we also included control groups that were orally administered 0.3 mL of water per mouse and 0.3 mL of 0.5% Tween 80 per mouse. Moreover, two additional groups were treated orally with 1000 mg/kg of Sm and 400 mg/kg of PE, respectively, which corresponded to the highest doses tested for each agent. Administration of experimental and control treatments to the four groups was performed daily from day 10 to day 18 of pregnancy. We included three additional groups that were administered both the mutagen and the antimutagen; these mice were orally administered 100, 500, and 1000 mg/kg of Sm from day 10 to day 18 of pregnancy, and with an intraperitoneal (ip) injection of HU (300 mg/kg) on day 16 of the assay. Similarly, three other groups were orally administered 100, 200, and 400 mg/kg of PE (from day 10 to 18) and given ip injections of HU (300 mg/kg) on day 16. Finally, we also included a positive control group that was given an ip injection of HU (300 mg/kg) on day 16.

At day 19 of the assay, a blood sample from the tail of each mother was obtained and spread onto two clean slides, fixed with methanol for 5 min, and stained with 10% Giemsa in PBS for 15 min. Additionally, the pregnant individuals were immediately dissected to obtain the fetuses, which were then cleaned, sexed, and organized for the teratogenic study, which was reported elsewhere. Moreover, the neck of each of the six fetuses was cut, and the obtained blood was spread onto two clean slides per individual. Subsequently, the cells on each slide were fixed for 5 min in methanol and stained for 15 min with Giemsa stain in PBS.

Micronuclei and bone marrow cytotoxicity

To evaluate the genotoxic effect of HU as well as the protection exerted by the tested antimutagens, we utilized the micronucleus test.¹⁶ Micronuclei were quantified as follows: the number of micronucleated polychromatic erythrocytes (MNPE) was determined out of 1000 polychromatic erythrocytes per individual, and the total number of micronucleated erythrocytes was determined in 10,000 erythrocytes per individual. Moreover, to examine possible agent-induced modifications on erythropoiesis induction, we quantified the number of polychromatic erythrocytes in 1000 erythrocytes per individual. Statistical analysis of the obtained data was performed with analysis of variance and Tukey tests using Sigma Stat Software (version 2.03).

Results

The results obtained with pregnant mice are shown in Table 1. We observed a low and homogeneous number of MNPE in the two control groups. The values determined for the highest doses of Sm and PE (1000 and 400 mg/kg, respectively), were in the same range as those of the control animals. However, the administration of HU led to a more than threefold increase in the mean amount of MNPE observed in the four groups mentioned above. In contrast, when the two highest doses of Sm were administered together with HU, the damage was significantly reduced compared with that induced by HU alone. The inhibitory effect observed in the presence of 1000 mg/kg of Sm was 53.4% of that obtained with HU alone.

Table 1.

Effect of Spirulina maxima and Its Protein Extract on the Number of Micronuclei and the Erythropoiesis Induced by Hydroxyurea in Pregnant mice

Compound	Dose (mg/kg)	MNPE Mean±SEM	MNE Mean±SEM	PET Mean±SEM
Water		12.6±1.3^a	77.8±3.8^a	124.2±3.9^a
Tween 80		10.4±2.0^a	83.3±5.4^a	122.5±4.5^a
Sm	1000	8.8±1.8^a	67.6±5.4^a	129.2±6.4^a
PE	400	9.3±2.1^a	72.5±9.1^a	128.5±5.3^a
HU	300	29.0±2.7^b	322.2±21.0^b	85.2±7.3^b
Sm+HU	100+300	26.4±4.5^b	283.4±25.3^b	98.7±10.2^b
Sm+HU	500+300	20.3±4.3^{a b}	244.9±35.6^{a b}	105.8±8.4^a
Sm+HU	1000+300	13.5±6.3^{a b}	198.2±28.9^{a b}	118.4±6.3^a
PE+HU	100+300	15.3±6.4^b	205.3±34.6^{a b}	117.9±7.9^a
PE+HU	200+300	25.2±7.2^b	267.4±42.7^{a b}	102.8±11.7^b
PE+HU	400+300	21.8±3.9^a	298.4±34.8^b	118.3±5.4^a

MNPE were scored in 1000 PET per mouse. MNE were scored in 10,000 erythrocytes per mouse. PET value was determined in 1000 erythrocytes per mouse. All determinations were made in five individuals per group. ANOVA and Tukey tests, P=.05.

Statistically significant difference with respect to the value obtained with 300 mg/kg of HU.

Statistically significant difference with respect to the mean of the two control values.

Sm, Spirulina maxima; PE, protein extract; MNPE, micronucleated polychromatic erythrocytes; PET, polychromatic erythrocytes; MNE, micronucleated erythrocytes; SEM, standard error of the mean; ANOVA, analysis of variance; HU, hydroxyurea.

Treatment with PE also caused significant decreases in MNPE at the highest tested dose (400 mg/kg) compared with that which was observed in mice exposed solely to HU (a decrease of 24.9%; Table 1). A similar situation was observed when we examined the micronuclei induced in all erythrocytes, that is, when we incorporated micronuclei present in both polychromatic and normochromatic erythrocytes. In the two control groups and in mice treated with Sm and PE alone, we noted a consistently small number of micronuclei, with no significant differences among the groups. With HU treatment alone, we noted an approximate fourfold increase over the number of micronuclei observed in the control groups. With two doses of PE plus HU or two doses of Sm plus HU, we noted a significant decrease in the number of micronuclei. Finally, with respect to the cytotoxic effect, the addition of HU caused a significant decrease (31%) in the amount of polychromatic erythrocytes compared with the two control treatments, which suggests that HU exerts a toxic effect on erythropoiesis. Additionally, a significant reversion of this damage was observed with two doses of Sm and two doses of PE.

The results obtained with the fetuses of the exposed pregnant mice are shown in Table 2. In this case, the numbers of MNPE were generally lower than those observed in the mothers; however, we also found that the numbers in the two control groups, as well as those in the groups administered Sm and PE, were similarly low with no significant differences among them. In contrast, the administration of HU induced a fivefold increase in micronuclei compared with the mean values of the two control groups. We observed a certain level of protection exerted by all tested doses of both antimutagenic agents, yet statistical significance was only achieved with the highest dose of PE. This result is most likely related to the variability observed in the individual data. The number of micronuclei in all erythrocytes reflected the behavior described above; however, in this case, the mutagen-induced increase in micronuclei was higher, with a >10-fold increase over the values observed in the two control groups. Additionally, the protection conferred by Sm and PE was more clearly evident. In this case, two doses of Sm and two doses of PE caused a statistically significant decrease when administered together with HU. The mean protection conferred by Sm was 33%, and the mean protection conferred by PE was 36.6%. Finally, with respect to bone marrow cytotoxicity, we observed a 15.2% decrease in the level of polychromatic erythrocytes induced by HU compared with the mean levels in the two controls; additionally, we found that Sm and PE were not significantly able to reduce such damage.

Table 2.

Effect of Spirulina maxima and Its Protein Extract on the Number of Micronuclei and the Erythropoiesis Induced by Hydroxyurea in the Fetuses of Treated Pregnant Mice

Compound	Dose (mg/kg)	MNPE Mean±SEM	MNE Mean±SEM	PET Mean±SEM
Water		7.1±1.3^a	29.3±7.3^a	825.4±70.2
Tween 80		6.3±0.9^a	30.4±10.1^a	837.4±100.5
Sm	1000	5.2±1.2^a	24.6±6.7^a	840.6±74.8
PE	400	4.5±1.4^a	27.8±8.4^a	832.3±57.9
HU	300	34.0±7.4^b	349.3±37.4^b	705.4±103.1
Sm+HU	100+300	22.1±6.3^b	275.4±50.4^b	765.3±45.6
Sm+HU	500+300	27.3±5.4^b	250.3±40.8^{a b}	758.1±48.9
Sm+HU	1000+300	28.2±7.3^b	217.9±48.1^{a b}	760±56.8
PE+HU	100+300	25.8±7.4^b	207.9±50.4^{a b}	743.1±62.5
PE+HU	200+300	27.3±6.2^b	235.2±30.8^{a b}	758.7±54.7
PE+HU	400+300	24.6±9.2^a	250.2±60.3^b	779.3±57.3

Statistically significant difference with respect to the value obtained with 300 mg/kg of HU.

Statistically significant difference with respect to the two control values.

Discussion

In the present report, we demonstrated the genotoxic effect induced by HU in both pregnant mice and their fetuses. This effect was evidenced by a significant increase in the number of MNPE and in the number of micronuclei per total number of erythrocytes. Therefore, HU acted as a chromosome-damaging compound in both adult mice and transplacentally.

HU is a chemical known to inactivate ribonucleotide reductase, thereby blocking DNA synthesis and inducing cell death; this process likely occurs through the participation of nitric oxide and intracellular peroxides. Therefore, HU is currently used to inhibit proliferation in patients with leukemia and some forms of neck and head malignancies.^17,18 In sickle cell anemia patients treated with HU, significant increases in the number of both micronucleated reticulocytes and micronucleated mature erythrocytes have been reported.¹⁹ Interestingly, various studies have determined that micronuclei induced by HU correspond to a particular subclass, which originates during the S phase and is related with the phosphorylation of Ser-139 of the histone H2AX, which is a marker for DNA double-strand breaks.^20,21 Thus, in this study, the micronuclei found in the exposed pregnant mice and their fetuses may belong to such a specific subclass. On the other hand, our study also uncovered the preventive effect of Sm and PE on the genotoxic and cytotoxic effects induced by HU. Overall, the protection was moderate (from 25% to 30% for both tested agents); however, this result reveals the participation of the extracted proteins in such protection. Additionally, our results suggest a specific role for these antimutagens in the inhibition of the molecular alteration related with this subclass of micronuclei and/or in the maintenance of an appropriate milieu for genomic stability; these roles are most likely related to the antioxidant properties of the antimutagens. Previously, it was reported that Sm decreased the number of micronuclei induced by the potent oxidative stress inducer, cadmium, in both pregnant mice and their progeny²²; thus, our present results extend the current knowledge regarding the anticlastogenic capacity of this alga and its constituents and show that this effect may be related with the prevention of distinct molecular damage.

Furthermore, our present data also indicate the relevance of the period in which the damage is induced and of the measured endpoint, as shown by the stronger protective effect found with the two agents with respect to congenital malformations¹⁰ compared with their anti-micronuclei potential. Our results illustrate the anti-micronuclei effect of the PE of Sm; however, we did not establish that this fraction provides the strongest antigenotoxic potential because Sm also showed a similar effect against HU and because a number of nonprotein compounds have been reported with high antigenotoxic or antioxidant properties, including vitamins, minerals, and carotenoids such as zeaxanthin.²³ Additionally, the lack of dose dependency, particularly with respect to the protein fraction, suggests the need of further studies to confirm such data and to analyze the relevance of the experimental conditions with respect to the intrinsic potency of the agent.

Studies examining micronuclei were initially carried out in hematologic cells, although other tissues have also been used over the years, such as oral, nasal, vaginal, or urinary cells. Additionally, transplacental observations can be carried out using umbilical cord cells or fetal tissues.¹¹ Therefore, knowledge regarding the placenta as a semipermeable barrier is increasing. It is known that almost any substance administered to the mother is capable of crossing the placenta to some extent, unless it is metabolized or altered during passage or if their molecular size and low lipid solubility do not allow the transfer. Studies on the matter have also revealed the variability in the kinetics of particular substances, which may depend on their particular chemical characteristics and may explain the differences in the time required to produce an effect. Additionally, other studies have established the participation of various transporters in the process, such as P-glycoprotein.^24,25 In this context, our results, as well as those of others, may highlight the importance of evaluating the transplacental genotoxic potential of substances; moreover, because numerous agents, such as antineoplastics, anticonvulsives, viruses, hormones, altered maternal nutrients, or malignant cells may cross the placenta and damage the fetus.^26,27 In addition, our results also demonstrate the importance of finding agents that may prevent or reduce the damage to the fetus as well as the importance of identifying adequate doses of chemopreventive or therapeutic agents in pregnant females. Incipient experimental studies in this field have been promoted in humans to evaluate the effects of chemotherapy, dioxin-like exposures, and maternal pathologies.^27

–30

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Campanella

, Crescentini

, Avino

. Chemical composition and nutritional evaluation of some natural and commercial food products based on Spirulina. Analysis, 1999; 27:533–540.

Mendes

, Nobre Cardoso

, Pereira

, Palavra

. Supercritical carbon dioxide extraction of compounds with pharmaceutical importance from microalgae. Inorg Chem Acta, 2003; 356:328–334.

Torres-Duran

, Ferreira-Hermosillo

, Juarez-Oropeza

. Antihyperlipemic and antihypertensive effects of Spirulina maxima in an open simple of Mexican population: A preliminary report. Lipids Health Dis, 2007; 6:33–39.

Miranda

, Cintra

, Barros

, Mancini Filho

. Antioxidant activity of the microalga Spirulina maxima. Braz J Med Biol Res, 1998; 31:1075–1079.

Ruiz-Flores

, Madrigal-Bujaidar

, Salazar

, Chamorro

. Anticlastogenic effect of Spirulina maxima extract on the micronuclei induced by maleic hydrazide in Tradescantia. Life Sci, 2003; 72:1345–1351.

, Ahn

, Kang

, Lee

. The effect of ultrasonicated extracts of Spirulina maxima on the anticancer activity. Mar Biotechnol, 2011; 13:205–214.

Romay

, González

, Ledón

, Ramirez

, Rimbau

. C-phycocyanin: A biliprotein with antioxidant, anti-inflammatory and neuroprotective effects. Curr Protein Pept Sci, 2003; 4:207–216.

Eriksen

. Production of phycocianin—a pigment with applications in biology, biotechnology, foods and medicine. Appl Microbiol Biotechnol, 2008; 80:1–14.

Bermejo-Bescós

, Piñero-Estrada

, Villar del Fresno

. Neuroprotection by Spirulina platensis protean extract and phycocyanin against iron-induced toxicity in SH-SY5Y neuroblastoma cells. Toxicol In Vitro, 2008; 22:1496–1502.

10.

Vázquez-Sánchez

, Ramón-Gallegos

, Mojica-Villegas

, Madrigal-Bujaidar

, Pérez-Pastén-Borja

, Chamorro-Cevallos

. Spirulina maxima and its protein extract protect against hydroxyurea-teratogenic insult in mice. Food Chem Toxicol, 2009; 47:2785–2789.

11.

Morita

, MacGregor

, Hayashi

. Micronucleus assay in rodent tissues other than bone marrow. Mutagenesis, 2011; 26:223–230.

12.

Qishen

, Guo

, Kolman

. Radioprotective effect of extract from Spirulina platensis in mouse bone marrow cells studied by using the micronucleus test. Toxicol Lett, 1989; 48:165–169.

13.

Premkumar

, Abraham

, Santhiya

, Ramesh

. Preotective effect of Spirulina fusiformis on chemical-induced genotoxicity in mice. Fitoterapia, 2004; 75:24–31.

14.

Boussiba

, Richmond

. Isolation and characterization of phycocyanins from blue-green algae Spirulina platensis. Arch Microbiol, 1979; 120:155–159.

15.

Warner

, Sadler

, Shockey

, Smith

. A comparison of the in vivo and in vitro response of mammalian embryos to a teratogenic insult. Toxicology, 1983; 28:271–282.

16.

Argüelles

, Álvarez-González

, Chamorro

, Madrigal-Bujaidar

. Protective effect of grapefruit juice on the teratogenic and genotoxic damage induced by cadmium in mice. J Med Food, 2012; 15:887–893.

17.

Bundy

, Marczin

, Chester

, Yacoub

. Differential regulation of DNA synthesis by nitric oxide and hydroxyurea in vascular smooth muscle cells. Am J Physiol, 1999; 277:H1799–H1807.

18.

Nagai

, Tarumoto

, Miyoshi

, Ohmine

, Muroi

, Komatsu

, Sassa

, Ozawa

. Oxidative stress is involved in hydroxyurea- induced erythroid differentiation. Br J Haematol, 2003; 121:657–661.

19.

Flanagan

, Howard

, Mortier

, Avlasevich

, Smeltzer

, Wu

, Dertinger

. Assessment of genotoxicity associated with hydroxyurea therapy in children with sickle cell anemia. Mutat Res, 2010; 698:38–42.

20.

, Sun

, Liu

, Guo

, Liu

, Jiang

, Zou

, Gong

, Tischfield

, Shao

. Replication stress induces micronuclei comprising of aggregated DNA double-strand breaks. PLoS One, 2011; 6:e18618.

21.

Wojtczak

, Poplonska

, Kwiatkowska

. Phosphorylation of H2AX histone as indirect evidence for double-stranded DNA breaks related to the exchange of nuclear proteins and chromatin remodeling in Chara vulgaris spermiogenesis. Protoplasma, 2008; 233:263–267.

22.

Chamorro-Cevallos

, Argüelles-Velázquez

, Álvarez-González

, Madrigal-Bujaidar

. Protective effect of Spirulina (Arthrospira) maxima against cadmium-induced genotoxicity, teratogenicity in mice. Proceedings 47th Congress of the European Societies of Toxicology (EUROTOX), Paris, France, 2011.

23.

, Wang

, Suter

, Rusell

, Grusak

, Wang

, Yin

, Tang

. Spirulina is an effective dietary source of zeaxanthin to humans. Br J Nutr, 2012; 108:611–619.

24.

Eshkoli

, Sheiner

, Ben-Zvi

, Holcberg

. Drug transport across the placenta. Curr Pharm Biotechnol, 2011; 12:707–714.

25.

Behravan

, Piquette-Miller

. Drug transport across the placenta, role of the ABC drug efflux transporters. Expert Opin Drug Metab Toxicol, 2007; 3:619–830.

26.

Tolar

, Neglia

. Transplacental and other routes of cancer transmission between individuals. J Pediatr Hematol Oncol, 2003; 25:430–434.

27.

Pedersen

, Halldorsson

, Mathiesen

, Mose

, Brouwer

, Hedegaard

, Loft

, Kleinjans

, Besselink

, Knudsen

. Dioxin-like exposures and effects on estrogenic and androgenic exposures and micronuclei frequency in mother-newborn pairs. Environ Int, 2010; 36:344–351.

28.

Desoye

, Gauster

, Wadsack

. Placental transport in pregnancy pathologies. Am J Clin Nutr, 2011; 96:1896S–1902S.

29.

Belkacemi

, Nelson

, Desai

, Ross

. Maternal undernutrition influences placental-fetal development. Biol Reprod, 2010; 83:325–331.

30.

Gwyn

. Children exposed to chemotherapy in utero. J Natl Cancer Inst Monogr, 2005; 2005:69–71.