Spermatogenesis Is Not Impaired in a Nucleotide Excision Repair–Deficient Min Mouse Model With or Without Neonatal Mutagen Treatment


  • Dr Ashild Andreassen,

    Corresponding author
    1. Department of Chemical Toxicology, Division of Environmental Medicine, Norwegian Institute of Public Health, Oslo, Norway
      Department of Infectious Diseases, Division of Infectious Diseases, Norwegian Institute of Public Health, PO Box 4404 Nydalen, NO-0403 Oslo, Norway (e-mail: ashild.andreassen@fhi.no).
    Search for more papers by this author
  • Inger-Lise Steffensen,

    1. Department of Food Safety and Nutrition, Division of Environmental Medicine, Norwegian Institute of Public Health, Oslo, Norway
    Search for more papers by this author
  • Ann-Karin Olsen,

    1. Department of Chemical Toxicology, Division of Environmental Medicine, Norwegian Institute of Public Health, Oslo, Norway
    Search for more papers by this author
  • Kiyoji Tanaka,

    1. Department of Frontier Biosciences, Graduate School of Frontier Biosciences, Osaka University and Solution-Oriented Research for Science and Technology, Japan Science and Technology Agency, Osaka, Japan
    Search for more papers by this author
  • Richard Wiger

    1. Department of Chemical Toxicology, Division of Environmental Medicine, Norwegian Institute of Public Health, Oslo, Norway
    Search for more papers by this author

Department of Infectious Diseases, Division of Infectious Diseases, Norwegian Institute of Public Health, PO Box 4404 Nydalen, NO-0403 Oslo, Norway (e-mail: ashild.andreassen@fhi.no).


ABSTRACT: Mice deficient in the xeroderma pigmentosum group A gene (Xpa) exhibit impaired nucleotide excision repair (NER) and are expected to accumulate bulky DNA adducts when subjected to certain compounds (eg, heterocyclic amines). Multiple intestinal neoplasia (Min) mice (B6Min/+) are particularly sensitive to low concentrations of mutagenic compounds in food. They develop intestinal tumors spontaneously, and the number and size of the tumors increase following exposure to 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), which humans are exposed to via fried food. We previously reported that NER is inefficient in adult testicular cells. Reduced NER (genetic deficiency; Xpa−/−) is expected to represent risk factors for PhIP-induced genotoxicity and could possibly disturb spermatogenesis, particularly in B6Min/+ mice. We therefore studied spermatogenesis in mice with combinations of Xpa and Min or wild-type genotypes 11 weeks after exposure to PhIP on days 3 to 6. Fewer offspring were obtained from B6Min/+Xpa−/− than from B6Min/+Xpa+/+ or B6Min/+Xpa+/−. Distributions of the different testicular cell types, indicative of normal spermatogenesis and relative testes weights, did not differ significantly in PhIP-exposed or unexposed mice regardless of their genotypes. We conclude that the removal of bulky DNA adducts does not seem to be essential for normal spermatogenesis.

The impact of exposure of gonocytes and early spermatogonial stem cells of neonatal mice to environmental chemical insults on subsequent somatic growth and gonadal development is important to elucidate. In 3- to 6-day-old neonatal mice, the male progenitor germ cells are present as gonocytes and spermatogonia A. Irreversible damage to these cells may have implications for the potential to generate mature spermatozoa and for the clones of spermatozoa to which they do give rise. The present study used 3- to 6-day-old mice with 3 different genotypes of xeroderma pigmentosum group A (Xpa), which encodes a key protein in nucleotide excision repair (NER). In addition, the mice were either wild type or heterozygous for the adenomatous polyposis coli (Apc) gene, which codes for a protein that interacts with actin and is involved in chromosome segregation and cell division. The mice were treated with a genotoxic concentration of the food mutagen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) to study possible effects of this DNA adduct–forming compound in combination with various genotypes on somatic growth, development of spermatogenesis, and relative growth of the testes and cauda epididymides.

The genome is protected from exposure to DNA-damaging agents by a variety of mechanisms, such as DNA repair. When such repair systems are defective, DNA damage induced by environmental agents may remain unrepaired, and this may result in subsequent DNA mutations. The integrity of the germ cell genome is of particular importance because it is the basis for the next generation. Sperm DNA damage is associated with reduced sperm quality, infertility, disturbance to embryo development, and early abortions (Fraga et al, 1991, 1996; Ni et al, 1997; Ahmadi and Ng, 1999a,b; Irvine, 2000; Shen and Ong, 2000; Loft et al, 2003; Schmid et al, 2003). Furthermore, paternal exposure to environmental agents such as cigarette smoke increases the risk of childhood leukemia and childhood malignant central nervous system tumors in the offspring, most likely via induction of DNA damage (Ji et al, 1997; Plichart et al, 2008; Lee et al, 2009).

To minimize the possibility of transferring DNA damage or germ-line mutations from parents to offspring, one expects that the cells of the germ line would exhibit effective DNA repair. However, in an earlier study, we observed that DNA adducts such as cyclobutane pyrimidine dimers (CPDs) induced after low doses of ultraviolet C (UVC) were not repaired in testicular cells from adult rats (Jansen et al, 2001). Similarly, Xu et al (2005) observed that male mice germ cells failed to remove CPDs. On the other hand, this same study reported approximately 60% removal of another UV lesion, 6-4 pyrimidone photoproducts, 12 hours following high doses of UVB irradiation, although at slower repair rates than in primary keratinocytes. We subsequently showed that other DNA adducts such as those induced by benzo(a)-pyrene (B[a]P) are poorly repaired in mouse male germ cells, such as spermatocytes and cells in the subsequent stages of spermatogenesis, because the DNA adducts persist in spermatozoa originating from the exposed precursor cells (Olsen et al, 2010). These results support the contention that male germ cells, at least in meiotic and postmeiotic cells, of the mature testicle in general display a reduced NER activity. Spermatogonia, on the other hand, seem to have functional NER. Spermatogonia have displayed low to moderate ability to repair CPDs (Xu et al, 2005). Similarly, B(a)P–DNA adducts induced in stem cell or differentiating spermatogonia did not persist in their resulting cohort of spermatozoa (Olsen et al, 2010).

Fried meat and fish contain heterocyclic amines (HCAs) such as PhIP, which has been shown to induce carcinomas of the mammary gland, colon, and prostate of rats and lymphoma in mice (Esumi et al, 1989; Ito et al, 1991; Rao et al, 1996; Shirai et al, 1997; Nagao, 1999). PhIP is usually the most abundant HCA in cooked meat (Felton et al, 1986; Kurosaka et al, 1992). A metabolite of PhIP covalently binds to the C-8 position of guanine, forming a DNA adduct (Frandsen et al, 1992; Nagaoka et al, 1992). DNA adduct formation is detected in somatic intestinal tissues, as well as in testis, after 1 single treatment with PhIP (Steffensen et al, 2006; Teunissen et al, 2010).

PhIP exposure induces mutations in various tissues from transgenic gptΔ mice and in Muta mice (Masumura et al, 1999; Chen et al, 2010). Previous studies on single-dose exposure of PhIP (5–50 mg/kg body weight [bw]) in B6Min/+ mice showed increased induction of intestinal tumors with loss of heterozygosity, chromosomal instability, and point mutations in intestinal tumors (Andreassen et al, 2002, 2006; Møllersen et al, 2004; Steffensen et al, 2006).

As would be expected in mice with defective DNA repair, several studies have observed that Xpa−/− mice treated with UV, PhIP, or B(a)P developed more site-specific tumors, and furthermore, exposure to PhIP led to higher levels of DNA adducts than similarly treated Xpa+/+ mice (van Steeg et al, 1998; Imaida et al, 2000; Klein et al, 2001; Steffensen et al, 2006).

Early studies using Xpa-deficient mice showed neither obvious physical abnormalities nor pathologic alterations, and the mice were fertile, producing viable offspring (Nakane et al, 1995). de Vries et al (1995) confirmed that Xpa-deficient mice showed normal fertility in their Xpa-deficient mouse. Furthermore, Apc-deficient mice seem to have an equivalent fertility to that of B6 wild-type mice. However, recently Nakane et al (2008) showed that aging Xpa-deficient mice (>6 months) exhibited impaired spermatogenesis and decreased testis weights.

B6Min/+Xpa−/− mice treated with 5 mg/kg bw PhIP had increased numbers of small intestinal tumors. PhIP–DNA adduct levels analyzed after exposure to 25 mg/kg bw PhIP were significantly increased in B6Min/+Xpa−/− mice compared with B6Min/+Xpa+/+ and B6Min/+Xpa+/− littermates. This shows that PhIP was metabolized and reached target organs (Steffensen et al, 2006). Knasmuller et al (1992) reported that PhIP treatment of adult mice induced PhIP–DNA adducts in testicular cells. Our hypothesis is that PhIP treatment leads to the formation of bulky DNA adducts, which, if they remain unrepaired, may interfere with DNA replication and cell division. This in turn could affect spermatogenesis and testis development. To our knowledge, effects of neonatal exposure on the development of spermatogenesis in these mouse strains, or in mice with the combined deficiency of both Apc and Xpa genes, have not been studied previously. As part of ongoing experiments studying the role of NER deficiency in tumor formation in neonatal mice treated with PhIP (Steffensen et al, 2006), we investigated possible effects on spermatogenesis. The majority of the germ cells in 3- to 6-day-old mice are progenitor cells (ie, gonocytes and type A spermatogonia) (Janca et al, 1986; McLean et al, 2003).

The present study was conducted to determine whether there are qualitative differences in spermatogenesis in mice with a mutated Apc gene in combination with deficiency of Xpa, compared with wild-type mice, and to determine if a single subcutaneous (SC) injection of 5 and 25 mg/kg bw of PhIP in neonatal mice exposing progenitor germ cells leads to disturbances in the development of spermatogenesis and testis weight. The study of genetic and treatment-related effects on progenitor cells of neonatal mice are of major importance also for risk assessment.

Materials and Methods

Animals and Housing

The Min pedigree was bred at the Norwegian Institute of Public Health, Oslo, Norway, by mating C57BL/6J-Apc+/+ (wild-type) (B6+/+) female mice with C57BL/6J-Apc+/Min (B6Min/+) males purchased from the Jackson Laboratory (Bar Harbor, Maine). The genotypes of all the animals were confirmed by allele-specific polymerase chain reaction (PCR) genotyping (Steffensen et al, 2006). Xpa gene–deficient mice were established by inserting a neomycin cassette sequence into exon 4 of the Xpa gene (Nakane et al, 1995) and later backcrossed onto a B6 background. The mice used in the present study represent a subgroup of a larger study concerning DNA adduct formation and the effects of PhIP exposure on intestinal tumorigenesis (Steffensen et al, 2006). The experiments reported in this article have been approved by the National Experimental Animal Board in Norway.

Homozygous mutant ApcMin/Min (Apc−/−) mice die during the embryonal stages (Moser et al, 1995), whereas homozygous Xpa−/− mice have been reported to have normal viability and fertility (Nakane et al, 1995). Therefore, 6 genotype combinations were obtainable from crosses between Min mice and Xpa gene–deficient mice. For further details, see Steffensen et al (2006).

The mice were housed in plastic cages with a 12-hour light/dark cycle, controlled humidity (55% ± 5%), and temperature (20–24°C). Water and diet were given ad libitum. The mice were given a breeding diet, SDS RM3 (E) during gestation and until weaning at 3 weeks and a standard maintenance diet, SDS RM1 (E) thereafter, both from Special Diets Services (Witham, United Kingdom).


PhIP (Chemical Abstracts Service 105650-23-5) of more than 98% purity was purchased from Toronto Research Chemicals (North York, Canada) was dissolved in concentrated HCl, which thereafter was evaporated. The PhIP-HCl was dissolved in 0.9% NaCl, and pH was adjusted to 3.5.

Treatment of Animals

Pups were given a single SC injection containing 0, 5, 25 mg/kg bw PhIP or 0.9% NaCl as vehicle control on days 3 to 6 after birth. All mice were sacrificed at 11 weeks.

Body and Organ Weights

The animals were weighed to the nearest 0.1 g following sacrifice. The testes and accessory sex organs were removed, and the testes were cleaned of excessive fat and connective tissue. These organs were blotted carefully on paper towels and weighed to the nearest mg. The cauda epididymides were removed using microsurgical scissors by cutting close to the junction between the corpus epididymis and the start of the vas deferens and weighed. The relative organ weights were calculated as follows: (sum of organ weight [mg]/body weight [g]) ×100.

DNA Flow Cytometric Analyses of Single Suspensions of Mouse Testicular Cells

One testis was transferred to a 60-mm Petri dish containing 1 to 2 mL of RPMI 1641 culture medium, the capsule was removed, and the testis was minced with curved surgical scissors to liberate individual cells. Cell suspensions were transferred to test tubes, and tissue fragments were allowed to settle for 1 minute. The supernatant was gravity filtered through a 55-μm nylon filter, and testicular cell samples were fixed in 0.1% paraformaldehyde in phosphate-buffered saline and analyzed immediately (Evenson et al, 1993; Oskam et al, 2004).

The DNA of testicular cells was stained by incubating 1 to 2 × 106 cells in RPMI 1640 medium containing 0.1% Triton X-100 and 1.0 μg of Hoechst 33258/mL for 15 minutes. Blue fluorescence was measured using an Argus 100 flow cytometer (Skatron, Lier, Norway). The percentages of cells in the 1C, 2C, S-phase, and 4C populations were estimated from DNA cytograms using the Multicycle program (Phoenix Flow System, San Diego, California).

Sperm Chromatin Structure Assay of Vas Deferens Sperm

The epididymides and vas deferens were quickly frozen on a block of dry ice following dissection. Immediately before flow cytometric (FCM) analyses, the vas deferens and some caput epididymides were quickly thawed in a 37°C water bath and transferred into 2 mL of ice-cold TNE buffer (0.15 M NaCl, 0.01 M Tris/HCl, 1 mM EDTA [pH 7.4]). Sperm were squeezed out of the vas deferens and expelled several times through a Pasteur pipette, filtered through 153-μm nylon mesh into test tubes, and kept on crushed ice until analysis by FCM, strictly following the procedure described by Evenson et al (2002). Abnormal chromatin structure, here defined as showing an increased susceptibility to acid denaturation, was measured in individual cells. When excited with blue laser light (488 nm), acridine orange intercalated into double-stranded DNA fluorescence green, this signal representing sperm with low levels of fragmented DNA. When associated with single-stranded nucleic acids, the red fluorescence represents sperm that have moderate to high levels of fragmented DNA, yielding a higher percentage of the DNA fragmentation index (DFI). The samples were analyzed on an EPICS XL flow cytometer (Beckman Coulter Inc, Miami, Florida). Recorded measurements were begun 3 minutes after staining, with a sampling rate of approximately 200 cells/second and a total of 1 × 104 cells processed for each sample.

Statistical Analysis

The data for spermatogenesis and relative testes weights were analyzed by Student's t test in addition to 1-way analysis of variance (ANOVA) (SigmaStat software, Jandel Scientific, Erkrath, Germany). When the 1-way ANOVA tests were not significant, the samples were tested in pairs by Student's t test (see footnote in Table 1). P < .05 was considered significant.

Table 1. . Effects of genotype and/or PhIP treatment on body weight, relative testes weight, and relative caudal epididymal weight in hybrid Apc × Xpa micea
GenotypebTreatment, mg/kg bwNo. of MiceMean Body Weight, g (SD)cMean Relative Testes Weight, mg/100 g bw (SD)Mean Relative Cauda Weight, mg/100 g bw (SD)
  1. Abbreviations: Apc, adenomatous polyposis coli; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; Xpa, xeroderma pigmentosum group A.

  2. a There were no significant differences within any of the tested variables by analysis of variance

  3. b When all of the samples were pooled according to Xpa genotype, there was a significant difference in body weight between Xpa−/− and the 2 other Xpa genotypes (P < .001)

  4. c This is the mean body weight of the subpopulation of males used in this study extracted from the larger study published by Steffensen at al (2006)

  5. d Statistical significant difference in body weight (P = .006) by Student's t test between these 2 groups

  6. e All animals died owing to toxicity

Apc+/+ Xpa+/+NaCl1027.25 (1.22)d657.76 (119.70)74.74 (7.39)
 5 mg PhIP527.90 (1.04)601.36 (158.92)73.34 (8.61)
 25 mg PhIP527.32 (1.07)737.58 (90,58)80.95 (3.42)
ApcMin/+ Xpa+/+NaCl525.33 (2.69)780.99 (60.77)75.66 (6.13)
 5 mg PhIP724.03 (3.43)790.66 (51.96)71.53 (13.29)
 25 mg PhIP523.66 (2.47)716,48 (141,49)67.59 (13.92)
Apc+/+ Xpa+/−NaCl1226.86 (1.08)731.90 (60.44)75.07 (11.40)
 5 mg PhIP1027.39 (1.65)682.16 (111.46)80.91 (13.29)
 25 mg PhIP528.61 (2.55)725.62 (83.49)77.77 (4.37)
ApcMin/+ Xpa+/−NaCl1025.90 (2.17)751.49 (68.71)79.66 (7.18)
 5 mg PhIP1025.04 (1.90)751.60 (123.14)78.84 (5.33)
 25 mg PhIP525.76 (0.45)732.18 (50,24)74.72 (5.18)
Apc+/+ Xpa−/−NaCl725.03 (1.68)c713.60 (76.67)66.84 (8.84)
 5 mg PhIP925.04 (2.33)649.12 (197.97)73.96 (10.40)
 25 mg PhIP0eee
ApcMin/+ Xpa−/−NaCl822.27 (3.38)754.15 (144.00)79.86 (18.82)
 5 mg PhIP923.73 (1.90)653.95 (225.21)75.90 (18.93)
 25 mg PhIP0eee


Body and Organ Weights in Relation to Genotype or Treatment

Changes in body weights and relative organ weights are good indicators of toxicity. The end points are widely used in guidelines for testing the effects of chemicals on male reproduction (European Union or Organisation for Economic Co-operation and Development guidelines). Changes in testis or epididymal weights are valid indicators for effects on spermatogenesis. As can be seen in Table 1, all groups consisted of 5 to 12 mice and were thus suitable for statistical analysis.

No changes were observed in body weights, relative testis weights, or relative cauda epididymal weights in relation to combined Xpa/Apc genotypes or PhIP treatment (Table 1). We further investigated whether there were differences in the body weights related to the Xpa zygosity alone, regardless of PhIP treatment or Apc status. The mean body weights ± SD for mice were as follows: Xpa−/−, 24.0 ± 2.56 g (n = 33); Xpa+/+, 26.0 ± 2.61 g (n = 37); Xpa+/−, 26.5 ± 2.08 g (n = 52). One-way ANOVA indicated that when we pooled all mice according to their Xpa genotype, the Xpa−/− mice had significantly lower body weights than both Xpa+/+ or Xpa+/− animals (P < .001). And by comparing the 2 genotypes, Apc+/+ Xpa+/+ vs Apc+/+ Xpa−/− exposed to vehicle, there was a statistical significant difference of P = .006 by Student's t test. Similar analyses for the relative testis weights and relative cauda epididymal weights showed no differences associated with Xpa genotypes.

DNA FCM Analysis of Spermatogenesis

FCM evaluation of cell distributions of primary testicular cell cultures has been shown to be a sensitive method to detect disturbances in spermatogenesis, both in humans and experimental animals (Evenson et al, 1993; Wiger et al, 1995). The mice in the present study all displayed normal spermatogenesis, and the variations between the groups were small and not statistically significant (Table 2).

Table 2. . Effects of genotype and/or PhIP treatment on spermatogenesis in hybrid Apc × Xpa mice measured by flow cytometry
GenotypeTreatment, mg/kg bwNo. of MiceMean Percent 1C (SD)Mean Percent 2C (SD)Mean Percent S-Phase (SD)Mean Percent 4C (SD)
  1. Abbreviations: Apc, adenomatous polyposis coli; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; Xpa, xeroderma pigmentosum group A.

  2. a All animals died owing to toxicity

Apc+/+ Xpa+/+NaCl1076.09 (3.19)11.43 (1.90)2.94 (1.20)9.52 (2.38)
 5 mg PhIP575.74 (4.10)10.16 (1.57)1.68 (0.37)12.40 (2.33)
 25 mg PhIP574.38 (4.30)11.16 (1.51)3.36 (1.33)11.08 (2.06)
ApcMin/+ Xpa+/+NaCl572.02 (3.79)12.6 (1.84)3.36 (1.55)11.64 (0.66)
 5 mg PhIP778,40 (3,39)8.43 (1.52)1.84 (0.74)11.33 (1.89)
 25 mg PhIP572.20 (3.82)12.34 (2.64)3.54 (1.14)11.90 (1.00)
Apc+/+ Xpa+/−NaCl1274.23 (3.25)12.16 (1.89)3.25 (1.36)10.40 (2.05)
 5 mg PhIP1077,42 (4,48)10.35 (2.56)2.26 (0.67)9.98 (2.11)
 25 mg PhIP571.54 (2.50)12.32 (0.94)4.00 (0.90)12.14 (0.98)
ApcMin/+ Xpa+/−NaCl1076.63 (3.85)11.58 (1.56)3.58 (1.21)8.18 (3.17)
 5 mg PhIP1079.93 (3.72)8.69 (2.27)2.09 (0.72)9.49 (1.41)
 25 mg PhIP573.74 (6.06)11.10 (2.20)3.62 (1.36)11.54 (2.65)
Apc+/+ Xpa−/−NaCl774.77 (1.69)11.73 (1.31)2.16 (1.05)11.43 (1.47)
 5 mg PhIP975.53 (3.29)11.37 (2.03)3.19 (1.17)9.91 (1.92)
 25 mg PhIP0aaaa
ApcMin/+ Xpa−/−NaCl873.44 (4.68)12.78 (3.67)2.73 (1.01)11.05 (089)
 5 mg PhIP976.81 (6.01)10.8 (3.80)2.69 (1.73)9.69 (1.70)
 25 mg PhIP0aaaa

Sperm Chromatin Structure Assay

The sperm chromatin structure assay (SCSA) is a method to detect DNA fragmentation and disturbances in the packaging of the male genome in mature mouse sperm cells and has been used as a sensitive end point in several toxicologic studies. We analyzed samples of vas deferens sperm from the various groups and found that the DFIs were generally very low, ranging from 0.63 to 1.96. The DFI of less condensed caput sperm is naturally much higher than that for vas deferens sperm. This indicates that the chromatin structure is unchanged in the analyzed samples (Table 3).

Table 3. . Studies of SCSA from mice of various genotypes
GenotypeOrganMeasured SCSASCSA Mean
  1. Abbreviations: Apc, adenomatous polyposis coli; SCSA, sperm chromatin structure assay; Xpa, xeroderma pigmentosum group A.

  2. a Sperm from the caput epididymis have more loosely packaged chromatin than either cauda or vas deferens sperm and thus reveal higher SCSA scores

Apc+/+ Xpa−/−Vas deferens0.730.63
 Vas deferens0.54 
ApcMin/+ Xpa−/−Caputa9.810.78
ApcMin/+ Xpa+/+Caputa19.6421.31
 Vas deferens1.851.96
 Vas deferens2.07 
Apc+/+ Xpa+/+Vas deferens1.821.64
 Vas deferens1.45 
 Vas deferens0.960.9
 Vas deferens0.63 
 Vas deferens1.1 
 Vas deferens0.781.49
 Vas deferens2.2 


NER is a thoroughly studied DNA repair pathway and is involved in the removal of a broad range of DNA lesions, mainly DNA adducts that disturb the DNA double-helix conformation (Hoeijmakers and Bootsma, 1990; Lehmann, 1995; Naegeli, 1995; Shuck et al, 2008). In the study by Steffensen et al (2006), PhIP–DNA adduct formation was detected in the small intestine, colon, and liver, and the number of small intestinal tumors was increased after treatment of 3- to 6-day-old mice with a single SC injection of 25 mg/kg PhIP. PhIP–DNA adducts were not measured after exposure to 5 mg/kg PhIP; however, because this dose increased the number of small intestinal tumors (Steffensen et al, 2006), it most probably gave rise to DNA adducts in the testis. Consequently, 5 mg/kg PhIP is considered a genotoxic dose. It has been shown in earlier studies that PhIP induces bulky DNA adducts in testicular tissue of adult mice and rats at 2.5 to 40 mg/kg bw and 50 mg/kg bw, respectively (Knasmuller et al, 1992; Huber et al, 1997). Similarly, in vivo exposure to other environmental agents such as B(a)P has been shown to give rise to DNA adducts in adult male testicular cells residing within the testis barrier (Olsen et al, 2010; Verhofstad et al, 2010). In the present study, the mice received a single SC treatment with PhIP on postnatal days 3 to 6, a period when the male germ cells present in the testis consist of gonocytes and spermatogonia type A (Janca et al, 1986; McLean et al, 2003) residing in the basement compartment of the seminiferous tubules. On day 3, the germ cells in the seminiferous tubules consist of gonocytes; on day 6, the gonocytes have differentiated into type A spermatogonia (Janca et al, 1986; McLean et al, 2003). Because of the localization of the cells and the previous observation that PhIP induces DNA adducts in the adult testis, it is likely that PhIP gave rise to DNA adducts in the testes of these neonatal mice, although this was not confirmed in this study. However, PhIP–DNA adducts have been detected after 1 single SC injection of PhIP both in somatic cells and in the testis (Steffensen et al, 2006; Teunissen et al, 2010). A more chronic exposure during the entire period of spermatogenesis is more likely to give rise to negative effects in the testes, similar to what is observed for male germ cell mutagenesis by B(a)P and PhIP (Steffensen et al, 2001; Olsen et al, 2010).

The lack of observed effects from a single dose of PhIP might be a result of the low-exposure dose and the fact that this was only a single dose injection in 3- to 6-day-old male mice. The exposure level may have been too low to cause observable effects in spermatogonia or gonocytes. At a higher dose, 25 mg/kg bw, most of the B6+/+Xpa−/− mice died. Another plausible explanation for the lack of effects after PhIP exposure is that the spermatogonia or gonocytes might be more resistant or more efficient to eliminate challenged cells than testicular cells at later stages of spermatogenesis.

Our group has previously shown that NER is suppressed or nonfunctional in adult rat testicular cells for some types of DNA adducts because of impaired incision (Brunborg et al, 1995; Jansen et al, 2001), as was also shown in mouse testicular cells, at least for CPDs (Xu et al, 2005). The Xpa protein, which is involved in the verification of DNA lesions, is expressed at very low levels in testicular germ cell tumor cells (Köberle et al, 1999).

A number of proteins required for DNA excision repair may also have additional functions besides their roles in excision repair. Thus, deficiency in different repair proteins may result in variable effects on development and survival of mice. Recently, a number of studies on repair-deficient mice have revealed that some of the proteins involved in the NER pathway are required for the health and survival of the individual. Xpg-deficient mice exhibited postnatal growth failure and died prematurely when they were approximately 3 weeks (Harada et al, 1999). Xpg is involved in the incision step during NER. Another DNA repair gene, Ercc1, also involved in the incision step of NER, is essential for normal spermatogenesis and oogenesis in mice (Hsia et al, 2003); however, mice deficient in Ercc1 otherwise grow normally. This seems not to be the effect of reduced levels or complete lack of Xpa protein because results from previous studies by Nakane et al (1995) and de Vries et al (1995) revealed that Xpa−/− mice had normal fertility. However, in the study by Steffensen et al (2006), it was shown that in the control group given 0.9% NaCl, fewer B6Min/+ Xpa−/− mice were born compared with the other 2 Xpa genotypes. This could indicate that the combination of reduced Apc protein and lack of Xpa has an effect on the development of offspring.

In the 11-week-old male mice, the distribution of testicular cells was considered normal in both the Xpa−/− and the B6Min/+ genotypes, indicating that neither of these genotypes alone or in combination had any significant effects on the development and progression of spermatogenesis, with respect to composition of testicular cells. We did not expect to observe any great differences in spermatogenesis associated with the 3 Xpa genotypes because we earlier reported lowered NER activity in the testicular cells of rats (Olsen et al, 2003, 2005), and lower NER activity in testes than somatic cells has also been reported for mice (Xu et al, 2005). Consequently, the treatment of all 3 genotypes with PhIP should theoretically have led to similar levels of PhIP adduct formation in the testicular germ cells. Apparently, spermatogenic cells tolerate a number of bulky adducts that are present even in ejaculated sperm. Bulky adducts of B[a]P diol epoxide adducts have been detected in human sperm, and the sperm with adducts were able to fertilize eggs in vitro and transfer the adducts to the conceptus (Zenzes et al, 1999a,b; Zenzes, 2000). Our latest data confirmed the detection of B[a]P diol epoxide–DNA adducts measured in sperm from mice exposed to B[a]P (Olsen et al, 2010). The induction of DNA damage by heavy smoking does not seem to have an impact on the capacity of spermatozoa to fertilize oocytes (Zenzes, 2000). However, early embryo development is apparently affected because smoking men are less successful than nonsmoking men with respect to pregnancy success rates, as exemplified by studies of couples undergoing assisted fertilization (Zitzmann et al, 2003). Furthermore, according to several studies, paternal cigarette smoking might have significant consequences on growth and development of the offspring, who also exhibit a significant increase in the incidence of childhood cancers (Ji et al, 1997; Plichart et al, 2008; Lee et al, 2009).

During the first 11 weeks, all mice displayed normal growth, and PhIP treatment did not seem to interfere with the growth process. However, when we examined the effect of Xpa zygosity alone on growth rate, we observed that our subpopulation of male Xpa−/− mice had significantly lower body weights than the Xpa+/+ or Xpa+/− mice. This finding is similar to and confirms the effects of genotype on body weight reported in Xpa−/− mice (Nakane et al, 2008). With regard to spermatogenesis, FCM analysis revealed that all groups displayed normal compositions of testicular cells. Neither PhIP treatment nor genotype interfered with spermatogenesis in up to 11-week-old mice. The lack of differences in the relative testes or cauda epididymal weight changes also indicates that neither PhIP treatment nor genotype interfered with spermatogenesis. These findings are in agreement with Nakane et al (2008), who observed normal testis growth and development during the first 5 to 6 months in Xpa−/− mice. Only after this period did they observe that spermatogenesis showed signs of impairment in Xpa−/− mice.

The final aspect we examined in a selection of mice was related to the chromatin structure of the caput epididymal and vas deferens sperm using SCSA. This assay monitors the susceptibility of sperm chromatin DNA to acid-induced denaturation, as reflected in the DFI, and has proven to be a sensitive tool when studying fertility and reproductive toxicity. In our study, SCSA analyses indicated that the vas deferens sperm samples from treated mice had tightly packed chromatin with very low DNA fragmentation, which was similar to control values. The data indicate that neither the genotypes of Xpa and Apc nor PhIP treatment apparently had any effect on chromatin packaging in spermatozoa.

In conclusion, mice with the Xpa- and Apc-deficient genotypes, whether neonatally treated with PhIP or not, seem to display normal spermatogenesis at 11 weeks, based on the methods used in this study. Apparently a single treatment regimen with 5 and 25 mg/kg bw PhIP did not affect the gonocytes or spermatogonia in such a way that the subsequent development of spermatogenesis was affected. Thus, the presence of Xpa or Apc in spermatogenic cells does not appear to be a requirement for normal testis growth or spermatogenesis, at least not during the first 3 months of life.

Neonatal exposure to a genotoxic environmental mutagen to which a large fraction of the population is exposed, or Xpa or Apc deficiencies, does not lead to disturbances of the spermatogenic process with respect to production of sperm in sexually mature mice. The presence of DNA adducts in the sperm, and their possible impact on the next generation, however, is an important question that remains to be elucidated.


We thank Dr Nur Duale and the laboratory apprentices Carl H. Fosli, Mathias Eriksen, and Tim R. Sterling for assistance with dissection and isolation of testes cells. We also thank Dr Gunnar Brunborg for valuable discussion. We are grateful to Dr Irma Oskam for performing the SCSA analysis and Prof Donald Evenson for teaching and allowing us to use SCSA to study sperm chromatin packaging.