Researchers in China are reporting that they have found a way to determine which somatic cells -- or differentiated body cells -- that have been reprogrammed into a primordial, embryonic-like state are the most viable for therapeutic applications.

In a paper published online last week by the Journal of Biological Chemistry, two collaborating teams from institutes at the Chinese Academy of Sciences point to a marker they found in induced-pluripotent stem cells, or iPS cells, taken from mice. That marker is a cluster of small RNA whose expression appears strictly correlated with levels of pluripotency, or "stemness." (The more pluripotent, the more likely a stem cell will develop into the desired tissue, organ or being.)

"We identified a genomic region encoding several genes and a large cluster of microRNAs in the mouse genome whose expression is high in fully pluripotent embryonic stem cells and iPS cells but significantly reduced in partially pluripotent iPS cells, indicating that the Dlk1-Dio3 region may serve as a marker," said Qi Zhou, a researcher at the CAS Institute of Zoology and co-author of the paper. "No other genomic regions were found to exhibit such clear expression changes between cell lines with different pluripotent levels."

After the creation of the first iPS cells in Japan in 2006, Zhou and others set out to determine whether the reprogrammed adult cells are versatile enough to generate an entire mammalian body, as embryonic stem cells can.

Then, last summer, Zhou announced that his team had reprogrammed somatic cells of mice, injected them into embryos and created 27 live offspring, which clearly demonstrated that iPS cells can, like embryonic stem cells, produce healthy adults. Though lauded as a huge step forward, they also found not all iPS cells were perfect: Many of the iPS cell lines used did not produce mice, and some of the mice that were produced had abnormalities.

"The success rate of obtaining iPS cells with full pluripotency was still extremely low, which significantly hindered the application of iPS cells in therapeutics and other aspects," Zhou said.

Believing that there might be some intrinsic gene expression difference between the lines of iPS cells with varying levels of pluripotency that could be identified at early culture stages, so that less viable lines could be abandoned and more viable lines focused on, Zhou teamed up with bioinformatics specialist Xiu-Jie Wang, who works at the Chinese academy's Institute of Genetics and Developmental Biology.

Together, their groups profiled the small RNA expression patterns of ES and iPS cell lines from different genetic backgrounds and with different pluripotent levels using Solexa technology.

"There are nearly 50 miRNAs encoded in this region, and those expressed miRNAs all exhibited consistent and significant expression differences between stem-cell lines with different pluripotency levels," Wang said. "With this discovery, iPS cells with different pluripotency can be distinguished in their early phases, which will, thus, significantly improve the production of full pluripotent iPS cells and promote their application in disease therapy," Wang said.

As stem cells can be applied in the treatment of many diseases related to tissue replacement or organ implantation, Zhou said, if the team's findings also are true for humans, "it will cause a revolution in stem-cell research and the application of it in the very near future."

        

Researchers from Universite Laval's Faculty of Medicine and the CHUQ Research Center have proven that it is possible to repair the defective gene responsible for Duchenne muscular dystrophy. The team, led by Professor Jacques P. Tremblay, is presenting its new therapeutic approach in an article published in the online version of the scientific journalGene Therapy.

Duchenne muscular dystrophy is a hereditary disease affecting one in 3,500 males. It causes progressive muscle degeneration that begins in early childhood and causes death by age 25 in most people afflicted. The disease is caused by mutations that affect a protein called "dystrophin." The mutations alter the normal nucleotide sequences of this protein's gene and stop its synthesis.

Professor Tremblay's team partnered with Cellectis, a French firm specializing in genome engineering, in order to design enzymes - called meganucleases - with the ability to correct the dystrophin gene. During in vitro testing, the researchers inserted genes coding for a variety of meganucleases into human muscle cells. They repeated the experiment in vivo with mice carrying the mutation that causes the illness. Both series of testing showed that the meganucleases can lead to a restoration of the normal nucleotide sequences of the dystrophin gene and its expression in muscle cells.

A number of hurdles must be overcome before this approach can be tested in humans, cautions Dr. Tremblay. "It must first be proven in laboratory animals that it is possible to insert a meganuclease targeting the dystrophin gene directly into muscle cells, and that this will induce the synthesis of dystrophin able to attach to the muscle fiber membrane," explains the researcher. "We're still two to three years away from this stage," he estimates. "Subsequent stages, including human trials, could take even longer," adds Dr. Tremblay.

        

Abstract

Background

Genomic control (GC) method is a useful tool to correct for the cryptic relatedness in population-based association studies. It was originally proposed for correcting for the variance inflation of Cochran-Armitage's additive trend test by using information from unlinked null markers, and was later generalized to be applicable to other tests with the additional requirement that the null markers are matched with the candidate marker in allele frequencies. However, matching allele frequencies limits the number of available null markers and thus limits the applicability of the GC method. On the other hand, errors in genotype/allele frequencies may cause further bias and variance inflation and thereby aggravate the effect of GC correction.

Results

In this paper, we propose a regression-based GC method using null markers that are not necessarily matched in allele frequencies with the candidate marker. Variation of allele frequencies of the null markers is adjusted by a regression method.

Conclusion

The proposed method can be readily applied to the Cochran-Armitage's trend tests other than the additive trend test, the Pearson's chi-square test and other robust efficiency tests. Simulation results show that the proposed method is effective in controlling type I error in the presence of population substructure.

Background

Population-based genetic association analysis is a powerful method for detecting susceptibility loci for complex diseases. A common issue in such design is that it may be subject to population heterogeneity and, as a result, spurious association may be reported if the population substructure is not properly addressed. Many methods have been proposed to deal with population heterogeneity in genetic association analysis.

When there is population stratification (PS) on allele frequencies, a direct method is to use family-based design [1-5] in which unaffected family members are chosen to match each case so that the association detected is truly due to the linkage between the candidate marker and the disease. But this method is limited by the cost and the difficulty in recruiting family members. Pritchard et al. [6,7] used a Bayesian clustering method to infer the number of subpopulations and to assign the individuals to putative subpopulations. The inferred memberships in each subpopulation are then used to perform tests of association for that subpopulation. A modification of this method was implemented by Satten et al. [8], in which subpopulation memberships were decided by a latent class model. Patterson et al. [9] proposed a principle components analysis method to correct for the population structure and obtained a test statistic based on the eigenvalues of the correlation matrix to detect the population structure. When the population has substructure, the usual chi-square statistics have non-central chi-square distributions under the null. Gorroochurn et al. [10] proposed a δ-centralization method to correct for PS by centralizing the test statistics using information from the null markers.

Another form of population heterogeneity is the cryptic relatedness or correlation across individuals. For this type of data, Devlin and Roeder [11] developed the genomic-control (GC) method to correct for the variance inflation. They proposed to use the additive Cochran-Armitage trend test to detect the gene-phenotype association. Assuming that the correlations or kinship coefficients are the same across all markers, they showed that the scaled test statistic has asymptotically a 1-df chi-square distribution. The scaling factor, known as the variance inflation factor (VIF), can be estimated from information of the unlinked null markers.

The GC method is a simple and effective method in association studies to correct for population heterogeneity caused by cryptic relatedness. However, when the GC method is applied to recessive or dominant trend tests [12] or, to Pearson's chi-square test [13] or other robust tests [14], the null loci are required to match with the candidate loci in allele frequencies, which reduces the number of available null makers.

In this study, we propose a regression-based genomic control (RGC) method that can be applied to association tests other than the additive trend test. This method allows for using arbitrary null markers in the GC correction procedure by adjusting the variability of the allele frequencies of the null markers through linear regression. We use simulation studies to check whether the method appropriately corrects for the problem of spurious association. In addition the robustness of the proposed method to the errors in selecting null markers is assessed. We also simulate the power of our method.

Methods

Trend tests

Let A be the high-risk candidate allele with the allele frequency p and a the normal one with the allele frequency q = 1 - p. To detect the association between the marker A and a disease, we assume that there are n₀ cases and n₁ controls with total n = n₀ + n₁ individuals. The genotype data are summarized in Table 1.

Denote the three genotypes by G₀ = aa, G₁ = Aa and G₂ = AA. Let f_i = P(case|G_i) be the penetrance given genotype G_i, i = 0, 1, 2. The null hypothesis of no association between the candidate marker and a disorder can be expressed as H₀: f₀ = f₁ = f₂. Since A is a high risk allele and a a normal one, let the score of genotype aa be 0, and that of AA be 1. For a specific choice of score x for genotype Aa, let

be the difference in weighted allele frequency between cases and controls, where = n_ij/n_i, i = 0, 1, j = 1, 2. When there is no allelic dependence or Hardy-Weinberg equilibrium holds in the population, Δ_x has, under the null hypothesis, the variance

(1)

where p₁ and p₂ are the frequencies of Aa and AA respectively. It can be estimated by

where = m_k/n is the estimate of p_k, k = 1, 2. The Cochran-Armitage's trend test indexed by x is then given by

(2)

In standard situation, has a central chi-square distribution with one degree of freedom under the null hypothesis. However, if there is cryptic relatedness, may be inflated. Denote the inflated variance of Δ_x under the null by = var_CR(Δ_x) and the variance inflation factor by . By this notation, in the presence of CR, Z_x ~ N(0, λ_x) under the null hypothesis. Illustrated in Figure 1 are the VIF λ₀, λ_0.5 and λ₁ as a function of the allele frequency p of the candidate marker. This figure was drawn from a simulated data of three subpopulations with (20, 30, 50) cases and (50, 30, 20) controls, and the Wright's coefficient F being 0.01. It shows that λ_0.5 of the additive model is a constant, λ₁ of a dominant trend test is a decreasing function of p while λ₀ of a recessive trend test is an increasing function of p. This verifies the results in [12].

The fact that the VIF λ_0.5 of the additive trend test doesn't depend on allele frequency of the candidate marker makes it possible that the λ_0.5 can be consistently estimated from a sequence of unlinked markers with arbitrary allele frequencies [11]. Unfortunately this is not true for trend tests with x other than 0.5 since the quantity λ_x does depend on allele frequency of the candidate marker. Therefore, dominant or recessive trend tests and other robust tests cannot be uniformly adjusted by the GC method using null markers with different allele frequencies. To overcome this problem, Zheng et al. [12] proposed to use null markers that have the same allele frequency as that of the candidate marker to evaluate the variance inflation factor. This constraint of matching allele frequency limits substantially the number of null markers that can be used.

RGC method

In what follows, we propose a regression-based GC method to adjust for the frequency variability of null markers when the GC method is applied to the general trend tests and the Pearson chi-square test.

In the Appendix, we show that when cryptic relatedness is present, under the null hypothesis the variance of Δ_x is a quartic polynomial of allele frequency p,

(3)

Gorroochurn et al. [10] pointed out that when the population has several subpopulations, Δ_x has a non-zero mean

(4)

When the population is of pure CR, the theoretical value of μ_x is zero. But in reality the PS and CR are usually mixed together, so it won't do any harm if we include this term in our analysis.

Let B₁, B₂, ..., B_K be arbitrary K null markers with minor allele frequencies p₁, ..., p_K. For the k-th marker, let be the genotype frequency estimate for genotype j in group i, i = 0, 1, j = 0, 1, 2. Then is the analogue of Δ_x for null marker B_k. Let be the sample estimate of p_k. Then we estimate the coefficients in (4) by minimizing

(5)

Denote the estimate of α_i by , i = 0, 1, 2. Let . The estimates of β_i, i = 1, 2, 3, 4 in (3) can be calculated by minimizing

(6)

Let p be the MAF of the candidate marker. Then we can estimate and μ_x by

The RGC-corrected Cochran-Armitage's trend test with score x can then be defined as

(7)

The Cochran-Armitage's trend tests are more powerful than the Pearson's chi-square test if the genetic model or x can be correctly specified. When the genetic model is unknown and the score x may be subject to misspecification, robust tests such as Pearson's chi-square test is preferred. Zheng et al. [13] proposed the following 2-df Pearson's chi-square test

(8)

where is the estimate of the correlation coefficient of Z₀ and Z₁. Combining (7) and (8) together, we therefore, propose the RGC-corrected Pearson's chi-square test as

(9)

Simulation study

To assess the validity of the proposed RGC method, we have implemented extensive simulations. Following [11], we use Wright's coefficient F to measure the correlation due to CR. Since it is difficult to simulate pure CR data, following [11,12] and [14], we employ the following procedure to generate a CR population. Let p be the allele frequency of a marker. Assume that there are L subpopulations including a₁,⋯, a_L cases and b₁, ⋯, b_L controls. We first generate p₁, ..., p_L independently from the Beta distribution Beta((1 - F)p/F, (1 - F)(1 - p)/F). We then generate L subpopulations having allele frequency p₁, ..., p_L respectively, assuming that within each subpopulation Hardy-Weinberg equilibrium holds. Finally we mix the L subpopulations together. From long run, this mixed population would resemble a pure CR population.

The details of the data generation are as follows. We used two subpopulations in each of our simulation. First we chose an allele frequency p of a marker which could be either a candidate marker or a null marker. We generated each of p₁ and p₂ from the Beta distribution Beta((1 - F)p/F, (1 - F)(1 - p)/F). Let C₁, C₂ represent the two subpopulations. We calculated the probabilities P(G_i|C_j), i = 0, 1, 2, j = 1, 2 according to HWE. The disease prevalence k_j in subpopulation C_j was estimated by

We then calculated and , the probabilities of genotype G_i in cases and controls in subpopulation C_j, by

Next we drew independent genotype counts (a_0j, a_1j, a_2j) of cases and (b_0j, b_1j, b_2j) of controls from multinomial distributions Mul() and Mul() respectively. We then mixed (a_0j, a_1j, a_2j) and (b_0j, b_1j, b_2j) up to obtain a case-control data set given in Table 1, with and for i = 0, 1, 2.

With this method of generating data, we simulated the cases of p = 0.2 and 0.45 where p is the minor allele frequency of candidate marker. The frequencies of unlinked null markers were selected randomly with equal probability from [0.1, 0.5]. The data for the K null markers with the same penetrances f₀ = f₁ = f₂ and a candidate marker with different penetrances f₀, f₁, f₂ are independently generated. The number of replicates in each simulation was 10, 000. To avoid the instability of the linear regression, the predictors were centered before to be fitted into the regression [15].

Results

A regression-based genomic control (RGC) method is proposed and applied to association tests other than the additive trend test. This method allows for using arbitrary null markers in the GC correction procedure, in which the variability of the allele frequencies of the null markers is adjusted by linear regression. The method is assessed by extensive simulation results. In addition, the robustness of the proposed method to the errors in selecting null markers is evaluated. We also simulate the power of our method.

Table 2 provides simulated type I error results for the uncorrected, GC and RGC tests. It shows that the uncorrected trend tests have highly inflated type I error and the type I errors of GC-corrected test deviate from the nominal level 0.05 more or less. As can be seen from Table 2 the RGC tests yield almost all the simulated type I errors around 0.05. The only exceptions are when p = 0.2, K = 200 and F is either 0.01 or 0.02 the RGC-corrected T₀ test yields p-values 0.063 and 0.065 respectively. This is because T₀ uses the count of genotype AA only, therefore the sample size for this test is small.

Table 3 presents the simulated power of RGC-corrected tests. From this table, we see that the trend tests with the correct mode of inheritance have optimal power. The Pearson's chi-square test has less power but is very robust as to model specifications.

Selection of null markers is an important issue when applying the GC method. The null markers are presumably unlinked to the disease, but in practice some linked loci may be chosen as null markers. To investigate the influence of the inclusion of linked markers in the set of null markers, we allowed the markers to be linked to the disease with probability 2%. Table 4 shows that the linked markers have some effect on the type I error which varies across genetic models. But the RGC method still controls the type I error around the nominal level 0.05.

Discussion

Case-control design is useful in detecting genes related to complex disease. For a case-control sample, if there is population structure and cryptic relatedness, spurious association between disease and genotype can occur due to variance inflation in the statistical tests. The genomic control method proposed by Devlin and Roeder [11] is a simple and effective method for eliminating spurious results caused by cryptic relatedness.

However when applying the GC method to correct for inflation of type I error of general trend test or the Pearson's chi-square test, it is required that the null markers are matched with the candidate marker in allele frequencies. This matching limits the applicability of the GC method. In this paper we propose a RGC method to correct for the population stratification effects which allows for use of any null markers. To adjust for the variability of allele frequencies of the null markers we estimate the inflated variance τ_x and the noncentral parameter μ_x by linear regression. This RGC method can be applied to the Cochran-Armitage's trend tests other than the additive trend test, with arbitrary score, the Pearson genotype-based association test and other robust efficiency tests.

Simulation results show that the RGC method can properly correct for the inflation of type I error of trend tests or Pearson's chi-square test caused by cryptic relatedness in the population. It is observed that the RGC method is slightly conservative for recessive trend test and anti-conservative for dominant trend test when the minor allele frequency is close to 0. We think that this is due to the instability of linear regression near the boundary of MAF values.

Conclusion

Simulation studies show that the RGC method can effectively correct for the variance inflation caused by cryptic relatedness and is robust to inclusion of linked loci in the selection of null markers.

Authors' contributions

TY carried out the implementation of the regression method, conducted all simulations and wrote the initial draft of the manuscript. YY developed the regression method and proposed the project. BH designed the study and wrote the final version of the manuscript. All authors read and approved the manuscript.

Appendix

Here, we calculate the variance of Δ_x under population structure and various genetic models. Assume that case-control samples come from L subpopulations, which include a₁, ⋯, a_L cases and b₁, ⋯, b_L controls, respectively. Thus , and n₀ + n₁ = n. We also assume that individuals from different subpopulations are independent. For each subpopulation, the genotypic frequencies are described by

(10)

where p_i is the frequency of the allelic A_i. Let

Using the results from Devlin and Roeder [11] and Zheng et, al [12], we have

(11)

(12)

(13)

where p is the frequency of the allelic A.

Acknowledgements

BH and YY are supported by Chinese Natural Science Foundation and Chinese Academy of Science Grant. TY is supported by USTC Graduate Student Innovation Foundation. The authors thank three anonymous reviewers for their helpful comments and Yifan Yang for careful reading of the manuscript.

References

1. Spielman R, McGinnis R, Ewens W: Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM).

   Am J Hum Genet 1993, 52:506-516. PubMed Abstract | PubMed Central Full Text

   Return to text

2. Curtis D: Use of siblings as controls in case-control association studies.

   Ann Hum Genet 1997, 61:319-333. PubMed Abstract | Publisher Full Text

   Return to text

3. Gauderman W, Witte J, Thomas D: Family-based association studies.

   J Natl Cancer Inst Monogr 1999, 26:31-37. PubMed Abstract | Publisher Full Text

   Return to text

4. Li Z, Gail M, Pee D, Gastwirth J: Statistical properties of Teng and Risch's sibship type tests for detecting an association between disease and a candidate allele.

   Hum Hered 2002, 53:114-129. PubMed Abstract | Publisher Full Text

   Return to text

5. Li Z, Gastwirth J, Gail M: Power and related statistical properties of conditional likelihood score tests for association studies in nuclear families with parental genotypes.

   Ann Hum Genet 2005, 69:296-314. PubMed Abstract | Publisher Full Text

   Return to text

6. Pritchard J, Stephens M, Donnelly P: Inference of population structure using multilocus genotype data.

   Genetics 2000, 155:945-959. PubMed Abstract | PubMed Central Full Text

   Return to text

7. Pritchard J, Stephens M, Rosenberg N, Donnelly P: Association mapping in structured populations.

   Am J Hum Genet 2000, 67:170-181. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

   Return to text

8. Satten G, Flanders W, Yang Q: Accounting for unmeasured population substructure in case-control studies of genetic association using a novel latent-class model.

   Am J Hum Genet 2001, 68:466-477. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

   Return to text

9. Patterson N, Price A, Reich D: Population structure and eigenanalysis.

   PLoS Genet 2006, 2(12):e190. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

   Return to text

10. Gorroochurn P, Heiman G, Hodge S, Greenberg D: Centralizing the noncentral chi-square: a new method to correct for population stratification in genetic case-control association studies.

    Genet Epidemiol 2006, 30:277-289. PubMed Abstract | Publisher Full Text

    Return to text

11. Devlin B, Roeder K: Genomic control for association studies.

    Biometrics 1999, 55:997-1004. PubMed Abstract | Publisher Full Text

    Return to text

12. Zheng G, Freidlin B, Li Z, Gastwirth J: Genomic control for association studies under various genetic models.

    Biometrics 2005, 61:186-192. PubMed Abstract | Publisher Full Text

    Return to text

13. Zheng G, Freidlin B, Gastwirth J: Robust genomic control for association studies.

    Am J Hum Genet 2006, 78:350-356. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

14. Zang Y, Zhang H, Yang Y, Zheng G: Robust genomic control and robust delta centralization tests for case-control association studies.

    Hum Hered 2007, 63:187-195. PubMed Abstract | Publisher Full Text

    Return to text

15. Ryan TP: Modern Regression Methods. Wiley-Interscience: New York; 1996.

        

Abstract

Background

Peroxisome proliferator-activated receptor delta belongs to the nuclear receptor superfamily of ligand-inducible transcription factors. It is a key regulator of lipid metabolism. The peroxisome proliferator-activated receptor delta gene (PPARD) has been assigned to a region on porcine chromosome 7, which harbours a quantitative trait locus for backfat. Thus, PPARD is considered a functional and positional candidate gene for backfat thickness. The purpose of this study was to test this candidate gene hypothesis in a cross of breeds that were highly divergent in lipid deposition characteristics.

Results

Screening for genetic variation in porcine PPARD revealed only silent mutations. Nevertheless, significant associations between PPARD haplotypes and backfat thickness were observed in the F2 generation of the Mangalitsa × Piétrain cross as well as a commercial German Landrace population. Haplotype 5 is associated with increased backfat in F2 Mangalitsa × Piétrain pigs, whereas haplotype 4 is associated with lower backfat thickness in the German Landrace population. Haplotype 4 and 5 carry the same alleles at all but one SNP. Interestingly, the opposite effects of PPARD haplotypes 4 and 5 on backfat thickness are reflected by opposite effects of these two haplotypes on PPAR-δ mRNA levels. Haplotype 4 significantly increases PPAR-δ mRNA levels, whereas haplotype 5 decreases mRNA levels of PPAR-δ.

Conclusion

This study provides evidence for an association between PPARD and backfat thickness. The association is substantiated by mRNA quantification. Further studies are required to clarify, whether the observed associations are caused by PPARD or are the result of linkage disequilibrium with a causal variant in a neighbouring gene.

Background

In various pig populations genome wide studies have been carried out to map quantitative trait loci (QTL) and to develop genetic markers for breeding. This resulted in more than 400 QTL for fatness [1]. The numerous QTL studies revealed chromosomal regions repeatedly linked to fatness traits. Some of the most significant QTL for backfat were identified on porcine chromosome 7 [2-5]. Most studies report a paradox of lower backfat to be caused by the allele originating from the breed, usually Meishan, with the higher backfat mean [2,5-7]. In this QTL region the peroxisome proliferative activated receptor delta (PPAR-δ) gene was mapped [8,9]. PPAR-δ is involved in the regulation of lipid metabolism, energy balance and insulin sensitivity [10]. Therefore, PPARD is considered to be a functional as well as positional candidate gene for backfat thickness. A study of 74 porcine SNPs across 5 chromosomes, with the majority located in proximity to backfat QTL, revealed an association of two markers in PPARD with backfat thickness [11].

The metabolic and histochemical characterisation of fat and muscle tissue from pigs with the chromosome 7 QTL alleles from Meishan and Large White showed differences in adipocyte size and number in backfat as well as differences in the basal rate of glucose incorporation into lipids and activity of lipogenic enzymes [12]. Therefore, PPARD is considered to be a promising candidate gene for the observed QTL effect [12]. A microarray-based experiment aiming at the identification of differentially expressed genes between lean Piétrain and 'obese' German Landrace pigs revealed an up-regulation of PPARD in Piétrain [13]. This suggests a possible effect on lipid deposition and strengthens the hypothesis of PPARD being a candidate gene for fatness.

The objective of this study was to systematically screen the PPAR-δ gene for polymorphisms and carry out association studies with identified variants. Furthermore, mRNA expression of PPARD variants was analysed in liver.

Results

Gene structure, splice variants and their expression

The genomic structure of the porcine PPARD was previously unknown. Therefore, it had to be determined from the porcine RefSeq mRNA sequence that consists of eight exons [GenBank: NM_214152]. Genomic sequence data was obtained by sequencing a porcine BAC (PigE-255B24). The assembly of the BAC shot gun sequences resulted in two genomic contigs containing PPARD. Sequence data was submitted to gene bank [GenBank: EU169095]. The first contig comprises the putative promoter, exon 1, the complete intron 1, exon 2 and approximately 15.7 kb of intron 2 (Figure 1). The second contig covers the region from exon 3 to exon 8 and 12.2 kb of the region 3' of the last exon (Figure 1). The resulting PPARD mRNA sequence is complete. All introns follow the GT-AG rule. The protein-coding sequence starts in exon 3 and ends in exon 8. The derived amino acid sequence is 94.6% identical to the sequence of human PPAR-δ. The common PPAR-δ transcript includes all eight exons. In addition, a splice variant without exon 2 was detected. Both PPAR-δ splice variants seem to be ubiquitously expressed since they were detected in liver, lung, backfat, muscle, kidney, brain, spleen and heart tissue.

Polymorphisms and haplotypes

All eight exons of porcine PPARD, exon flanking intronic regions and 2000 bp of the putative promoter region were screened for genetic variation by resequencing the parental generation of a Mangalitsa × Piétrain intercross and three unrelated animals of each of the German Landrace, German Large White and Duroc breeds. A total of 25 variants were identified, comprising 22 SNPs, two insertion/deletion polymorphisms and one stretch of a variable number of cytosins (Table 1). Two out of 22 SNPs are located in the protein-coding region, but they do not change the amino acid sequence. The number of cytosins in the polyC stretch varied between 11 and 14 Cs in the analysed animals. However, it was impossible to determine the exact number of Cs of the polyC stretch in some heterozygous animals. For that reason, only 24 polymorphisms were used to infer haplotypes.

A total of five haplotypes were detected with haplotypes 4 and 5 being identical at 23 out of 24 polymorphisms (Table 2). Both founding Mangalitsa boars of the resource population were homozygous for haplotype 1. The haplotype frequencies were 17%, 46%, 17%, 8% and 12% for haplotype 1, 2, 3, 4 and 5, respectively, in the twelve Piétrain parental animals. Four SNPs (polymorphisms ss161109995, ss161109998, ss161110009 and ss161110010) are sufficient to tag the five haplotypes.

Association studies

Association between PPARD haplotypes and backfat thickness was initially studied in the F2 generation of a Mangalitsa × Piétrain cross. Analysis of haplotype frequencies in the F1 generation revealed that haplotype 4 was not passed on to the F1 generation and was consequently assumed to be absent in the F2 generation. Therefore, SNPs ss161109995, ss161109998 and ss161110010 were sufficient to distinguish the remaining four haplotypes and were used to genotype 599 F2 animals by diagnostic restriction enzyme assays. Association analyses of PPARD variants and backfat were carried out and revealed a significant association between PPARD haplotype 5 and backfat thickness (p = 0.022, Table 3) in the Mangalitsa × Piétrain cross when contrasted against all other haplotypes. However, the p-value is above the 5% bonferoni corrected significance threshold of 0.0125 that corrects for testing of four different haplotypes. Heterozygous animals carrying one haplotype 5 allele showed an increase in backfat thickness of 2.43 mm (Table 3).

This result was followed up in unrelated animals of a commercial pig breed. The selection of a suitable study population was problematical because haplotype 5 is infrequent in all analysed pig populations (Figure 2). The highest frequency was observed in Piétrain and is estimated at 5% (Figure 2). Nevertheless, haplotype 4 is identical to haplotype 5 at all but one SNP (ss161110009, Table 2) and haplotype 4 is frequent in German Landrace, German Large White and Duroc (Figure 2). German Landrace was chosen because it was expected to exhibit both a relatively high frequency of haplotype 4 and possibly a few animals carrying haplotype 5. A total of 681 animals were successfully genotyped. The frequencies were 10%, 44%, 7%, 37% and 2% for haplotypes 1, 2, 3, 4 and 5, respectively. Genotype frequency of all tag SNPs did not deviate from Hardy-Weinberg equilibrium. Backfat thickness was significantly decreased by PPARD haplotype 4 (p = 0.034) when tested against all other haplotypes. In contrast, haplotype 5 that differs only in one SNP from haplotype 4 increased backfat thickness in the Mangalitsa × Piétrain population (Table 3). Haplotype 5 has no significant effect on backfat thickness (p = 0.242) in the German Landrace population. The power to detect a significant association (p < 0.05) of the rare haplotype 5 (MAF = 2%) within the analysed German Landrace population of 700 individuals is only 0.05. Therefore, we cannot exclude an effect of haplotype 5 on backfat thickness in this population. The least square means of backfat thickness within the group of pigs carrying one haplotype 5 allele is higher than in pigs carrying no haplotype 5 (Table 3), which is the same trend as seen in the Mangalitsa × Piétrain population. Least square means of pigs homozygous for haplotype 5 cannot be reliably estimated since there are only two animals in that group.

Expression of PPARD variants

Our association studies suggest evidence for association between PPARD variants and backfat thickness. However, since two different haplotypes were associated with an opposing effect on backfat thickness in two different pig populations, it is unclear if the observed associations are caused by a variation in PPARD or a causal mutation in linkage disequilibrium to PPARD. None of the SNPs located in the coding region cause an amino acid exchange. For this reason, there is no obvious functional candidate for the observed association. However, numerous studies have identified cis-regulatory and synonymous mutations with functionally significant consequences for morphology, physiology and behaviour [14-16]. To estimate whether the detected polymorphisms in PPARD could be functional, gene expression of the PPARD haplotypes 4 and 5 was studied. Pigs containing the desired PPARD variants were bred by artificial insemination of F3 sows with Piétrain boars known to possess haplotype 4 and haplotype 5, respectively. This was necessary since none of the F3 animals possessed haplotypes 4 and 5. Consequently, offspring carrying no or one haplotype 4 or 5 allele was obtained. Tissue samples from liver were collected within 30 min after exsanguation and stored in RNAlater™ (Qiagen, Hilden, Germany). RNA was isolated, reverse transcribed and PPAR-δ expression was analysed in a relative quantification approach with Tata-Box binding protein (TBP) and Topoisomerase 2 beta (TOP2 as reference genes. PPAR-δ expression is significantly reduced by haplotype 5 and increased by haplotype 4 (Table 4). Interestingly, haplotypes 4 and 5 are the two haplotypes associated with backfat thickness. The presence of haplotype 5 increased backfat thickness in a Mangalitsa × Piétrain cross and the presence of haplotype 4 decreased backfat thickness in a German Landrace population. In accordance with these findings, PPAR-δ expression is altered in opposite directions by these two haplotypes.

Statistical significant reduced expression was reached by haplotype 1 as well. However, in the analysis absence of haplotype 1 was compared with mostly heterozygous presence of haplotype 1. Further analysis of PPAR-δ expression in animals with diplotypes containing haplotype 1 showed a reduced expression only in animals carrying diplotype 1/5, which is most likely caused by haplotype 5.

Discussion

PPARD was chosen as a candidate for backfat thickness because of the central role of PPAR-δ in the regulation of lipid metabolism and because of its localisation in a major QTL region for backfat thickness on chromosome 7. The candidate gene analysis of porcine PPARD presented here identified only silent mutations. Nonetheless, it reveals an association of PPARD haplotypes 4 and 5 with backfat thickness. Haplotype 5 is associated with increased backfat thickness in F2 Mangalitsa × Piétrain pigs, and haplotype 4 with decreased backfat thickness in the German Landrace population. It was not possible to carry out an association study with haplotypes 4 and 5 in the same pig population, as haplotype 4 is absent in the F2 Mangalitsa × Piétrain generation and haplotype 5 is extremely rare in German Landrace. Analogous to numerous QTL studies [2,5-7], the same paradox of lower backfat caused by the allele originating from the breed with more backfat was revealed. In the Mangalitsa × Piétrain cross, haplotype 5 originates from the lean Piétrain breed and causes higher backfat.

Interestingly, the opposite effect of PPARD haplotypes 4 and 5 on backfat thickness is reflected by an opposite effect of these two haplotypes on PPAR-δ mRNA levels. Haplotype 4 is associated with reduced backfat thickness, and it significantly increases mRNA expression of PPAR-δ in liver. Haplotype 5 is associated with higher backfat thickness, and it significantly decreases PPAR-δ expression in liver. These findings are in line with studies demonstrating a decrease of body fat in mice caused by PPAR-δ overexpression [17]. In conclusion, findings from the association study, when considered together with results from the PPAR-δ expression study may suggest an effect of PPARD variants on backfat thickness. However, this study was not able to verify that, because no obvious functional candidate was identified. None of the SNPs located in the coding sequence result in an amino acid exchange. None of the SNPs in the analysed 2000 bp region of the PPARD promoter are located in a conserved region or a region that is known to harbour a transcription factor binding site [18]. Differences in mRNA levels between PPARD haplotypes suggest that functionality is caused either by influences on mRNA stability or by differences in PPAR-δ mRNA de-novo synthesis. None of the SNPs located in mRNA exhibit a large influence on the mRNA secondary structure (data not shown). However, control elements located in introns or far away from the gene can enhance or inhibit mRNA expression. This makes it difficult to identify the functional variant, especially as the observed effect might be due to not only one, but several genetic variants interacting with each other. In conclusion, this study was not able to detect a genetic variant in PPARD that is likely to cause the observed association. Hence, the effect on backfat thickness can originate from another variant in linkage disequilibrium to the analysed haplotypes located either in PPARD or a neighbouring gene.

Conclusion

The candidate gene study involving PPARD revealed association between PPARD variants and backfat thickness. However, it is not clear whether or not the association is caused by PPARD or a neighbouring gene in linkage disequilibrium, especially since no obvious functional variant was identified. Further studies are required to determine whether the observed associations are present in other pig populations. Additionally, other candidate genes at the location of the QTL on porcine chromosome 7 should be considered.

Methods

Animals

Mangalitsa × Piétrain intercross

Initially, two Swallow Bellied Mangalitsa boars were crossed to 13 Piétrain sows to produce an F1 generation. All Piétrain sows were homozygous for the mutant Cys614 allele at the RYR1 locus. Selected F1 sows (n = 18) and boars (n = 5) were mated and the resulting F2 generation was used for the analysis presented here. Animals were fed ad libitum. Male pigs were castrated. Pigs were slaughtered when they reached a weight of about 95 kg. Backfat thickness at the middle of the back was measured after slaughtering according to German Pig Breeders Standards [19]. A mean backfat thickness of 29.27 mm with a standard deviation of 5.57 mm was observed.

Association analyses were carried out in the F2 generation. F3 sows were backcrossed with Piétrain boars to introduce a desired PPARD gene variant (haplotypes 4 and 5) that was lost in F3. The genotypes of Piétrain boars were determined in order to assure the presence of the PPARD haplotypes 4 and 5 in the employed Piétrain boars. Seven F4 litters were used for gene expression studies. These seven litters correspond to a total number of 56 animals, 31 of them male and 25 female. They were slaughtered at 80 ± 2 days of age with an average weight of 25.6 kg. Tissue samples (approximately 0.5 g) for RNA isolation were collected from liver, longissimus muscle, backfat, heart, spleen, brain, kidney, ham and lung within 30 min after exsanguination and stored in 5 ml of RNAlater™ (Qiagen, Hilden, Germany).

Commercial Pig Breeds

A total of 722 German Landrace pigs raised under standardized condition for performance testing between 2002 and 2005 in Bavaria were used for association analysis. Backfat measurement was carried out after slaughtering according to German Pig Breeders Standards [19]. A mean backfat thickness of 20.64 mm with a standard deviation of 3.78 mm was observed.

The estimation of allele frequencies in different breeds was performed in 213 Piétrain, 40 German Landrace, 13 Large White, 24 Meishan and 45 Duroc animals.

BAC clone

A bacterial artificial chromosome (BAC) clone containing PPARD was identified in-silico with the help of the genomic location of the human gene from the Porcine BAC End Sequencing Project [20]. Colony PCRs with primers located in the putative promoter region and in exon 8 (primer pair 2 and 15) were carried out to ensure that the clone contains PPARD. Sequencing of BAC PigE-255B24 was performed by a shotgun approach as follows: Sheared fragments of 3 kb in length (GeneMachines) were subcloned separately into pUC19 vector. 4 × 384 clones were selected from the clone library. Plasmid DNA was prepared following a protocol supplied by Millipore (Schwalbach, Germany). Cycle sequencing was routinely performed using ABI PRISM BigDye Terminator v 3.1. Ready Reaction Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and M13f/M13r primer. All separations were run on ABI 3730XL capillary sequencers. Data were assembled and edited using the GAP4 program [21].

DNA-Isolation, primer design, PCR, Re-sequencing

DNA was isolated from blood, semen and ear tissue by standard methods. Primers were designed with Primer 3 Software [22] based on porcine GSS available through the NCBI homepage [23] and on derived BAC sequences. The primers used for Polymerase Chain Reaction (PCR) are summarized in Table 5. A standard PCR reaction contained 0.5 μM of each Primer, 200 μM of each dNTP (Fermentas, St. Leon-Rot, Germany), 0.5 U Taq-Polymerase (Qiagen), 50 ng genomic DNA and the diluted 10-fold PCR buffer supplied by Qiagen (Tris/HCl buffer (pH = 8.7) containing 15 mM MgCl₂, KCl, (NH₄)₂SO₄) in a total volume of 20 μL. After preincubation at 94°C for 3 min, the PCR mixture underwent 30 cycles of denaturation at 94°C for 30 s, annealing for 60 s and extension at 72°C for 60 s. A final elongation step at 72°C for 3 min followed. The annealing temperature was adjusted to the requirements of the primers (Table 5). Cleaning of PCR products was undertaken in a MultiScreen^® PCRμ96 Plate (Millipore). The amount of cleaned PCR product used for sequencing reaction varied from 2 to 5.5 μL depending on fragment size and PCR efficiency. In addition to the PCR products, the sequencing reaction consisted of 2 μL reaction mix of the BigDye^® Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems) and 0.5 μM primer. The volume of the sequencing reaction mix was adjusted to 10 μL. Thermal cycling for each primer was at 96°C for 10 s, 51°C for 5 s and 60°C for 4 min, for a total of 35 cycles. The sequencing reaction was cleaned via gel filtration with Sephadex G-50 (Sigma-Aldrich, Steinheim, Germany) in a MultiScreen^® 96 well filtration plate (Millipore). DNA sequencing was performed on an ABI 377 automated sequencer (Applied Biosystems) according to manufactures instructions. Obtained sequences were analyzed using Phred/Phrap/Polyphred/Consed software suite [24-27].

Genotyping

Genotyping of tag SNPs was performed by PCR followed by restriction enzyme assays. For RFLP analysis, 3 - 7 μl of the appropriate PCR product were mixed with the enzyme and the supplied buffer. The volume was adjusted to 10 μl using water. An overview of enzymes used and reaction conditions employed is given in Table 6. After incubation at 37°C the resulting fragments were separated on a 2% agarose gel.

Bioinformatics

Transcription factor binding sites were predicted by Cister [28], P-Match [29] and MatInspector [30]. Prediction of mRNA secondary structure was carried out using the Mfold web server [31].

Statistical analyses

Haplotypes were inferred using PHASE software version 2.1.1 [32,33] with default parameters (number of iterations = 100, thinning interval = 1, burn-in = 100).

Statistical analyses were carried out using the R environment for statistical computing version 2.4.1 [34]. Association between haplotypes and backfat thickness was estimated within 599 F2 Mangalitsa × Piétrain animals by a linear model with fixed effects of haplotype, dam and gender as well as covariate of living weight. For normalisation, backfat values were transformed to the power of 0.75. The model for estimating the effect of PPARD variants in the German Landrace population contained the performance testing station and weight as the only covariates because unrelated castrated animals were chosen. P-values were corrected for multiple testing of four haplotypes by Bonferroni correction. Least square (LS) means and their standard errors were calculated with untransformed backfat data based on the model described above using the effects package for R [35].

Gene expression studies

Total RNA from 20 mg of RNAlater™ (Qiagen) stabilised liver tissue was isolated using the RNeasy^® Plus Mini Kit (Qiagen). Homogenisation of the tissue was achieved by processing the sample in the FastPrep^® Instrument (Qbiogene, Inc, CA, USA) for 40 seconds at a speed setting of 6.5 m/s using Lysing Matrix D (Qbiogene). Synthesis of cDNA was carried out for all samples at the same time with 1 μg RNA and 500 ng random pentadecamer primers using the First Strand cDNA Synthesis Kit (Fermentas). Quantitative Real-Time PCR was carried out on an ABI PRISM^® 7000 Sequence Detection System (Applied Biosystems). Real-Time PCR reaction consisted of Power SYBR^® Green PCR Master Mix (Applied Biosystems), primers in an optimised concentration (PPARD: 100 nM 5'-CATGTCTCACAACGCCATTCG-3'/300 nM 5'-ATGTCGTGGATCACAAAGGGC-3'; TBP: 200 nM 5'-GATGGACGTTCGGTTTAGG-3'/300 nM 5'-AGCAGCACAGTACGAGCAA-3'; TOP2B: 200 nM 5'-GCTGGTGGCAAACACTCACTGG-3'/500 nM 5'-TGGAAAAACTCCGTATCTGTCTC-3') and diluted cDNA in a reaction volume of 20 μl.

After activation of Hot Start Polymerase by 10 min incubation at 95°C a 2-step PCR program was used consisting of 45 cycles of 15 s at 95°C and 1 min at 60°C. In case of PPARD annealing temperature had to be increased to 66°C to avoid primer dimers.

Crossing point (CP) and efficiency were calculated for each individual PCR reaction using ABI PRISM^® 7000 SDS Software (Applied Biosystems) and the MoBPA package in R [36], respectively. For statistical analysis a modified version of the REST^© (Relative Expression Software Tool) method was applied [37]. The algorithm of REST^© allows group-wise comparison of relative expression data in Real-Time PCR. However, this method assumes equal amplification efficiencies in all samples. The method used here was adapted to account for differences in PCR efficiency.

Mean expression differences between different PPARD genotypes for all genes of interest were calculated as follows. In a first step, the expression E for each animal and each gene was determined (Equation 1).

(1)

The geometric mean of these expression values was calculated within a group of N_gt animals with the same genotype gt (Equation 2).

(2)

Finally, the mean expression of one genotype (gt2) was divided by the mean expression of the other genotype (gt1). For SNPs where 3 genotypes were present in the analysed animals the expression of one genotype (gt1) was compared to a pool of animals with the two other genotypes (gt2). The mean expression difference R was calculated by dividing the ratio for the gene of interest by the geometric mean of the ratios for M reference genes (Equation 3).

(3)

A mean expression difference R = 1 characterises no effect of the genotype on expression of the analysed gene. The significance of a derivation from R = 1 was estimated by a permutation technique (number of permutations = 5000). The natural logarithm of R was used to obtain valid p-values for a two-sided significance test, because the untransformed values of R are left-skewed.

List of abbreviations

BAC: Bacterial artificial chromosome; GSS: Genome survey sequence; LS: Least square; MAF: Minor allele frequency; PCR: Polymerase chain reaction; RFLP: Restriction Fragment Length Polymorphism; SNP: Single nucleotide polymorphism; QTL: Quantitative trait locus.

Authors' contributions

KM carried out genetic studies, participated in statistical analysis and drafted the manuscript. HS participated in statistical analysis and edited the manuscript. RF designed the study and edited the manuscript. HB coordinated the BAC sequencing carried out by SS and MS. All authors read and approved the final manuscript.

Acknowledgements

We thank Bettina Hayn, Hermine Kienberger, Dagmar Reinl and Theresia Böhm for excellent technical assistance and Mahdi Osman for his contribution to data analysis. The expert animal care by the staff of WZW Experimental Station Thalhausen is greatly appreciated. We are grateful for specimen and DNA samples we received from Besamungsstation Bergheim e.V. and Bayrische Landesanstalt für Landwirtschaft, Grub (Germany). This project was supported by the Deutsche Forschungsgemeinschaft, Germany (DFG) grant FR 1284/2-1 to Ruedi Fries. Sequencing of the BAC clone was carried out within QuaLIPID, a project funded by the German Federal Ministry of Education and Research (grant 0313391D to Helmut Blöcker).

References

1. Hu Z-L, Dracheva S, Jang W, Maglott D, Bastiaansen J, Rothschild M, Reecy J: A QTL resource and comparison tool for pigs: PigQTLDB.

   Mammalian Genome 2005, 16(10):792. PubMed Abstract | Publisher Full Text

   Return to text

2. Rohrer GA, Keele JW: Identification of quantitative trait loci affecting carcass composition in swine: I. Fat deposition traits.

   J Anim Sci 1998, 76(9):2247-2254. PubMed Abstract | Publisher Full Text

   Return to text

3. Milan D, Bidanel JP, Iannuccelli N, Riquet J, Amigues Y, Gruand J, Le Roy P, Renard C, Chevalet C: Detection of quantitative trait loci for carcass composition traits in pigs.

   Genet Sel Evol 2002, 34(6):705-728. PubMed Abstract | BioMed Central Full Text | PubMed Central Full Text

   Return to text

4. Malek M, Dekkers JC, Lee HK, Baas TJ, Rothschild MF: A molecular genome scan analysis to identify chromosomal regions influencing economic traits in the pig. I. Growth and body composition.

   Mamm Genome 2001, 12(8):630-636. PubMed Abstract | Publisher Full Text

   Return to text

5. de Koning DJ, Janss LLG, Rattink AP, van Oers PAM, de Vries BJ, Groenen MAM, Poel JJ, de Groot PN, Brascamp EW, van Arendonk JAM: Detection of Quantitative Trait Loci for Backfat Thickness and Intramuscular Fat Content in Pigs (Sus scrofa).

   Genetics 1999, 152(4):1679-1690. PubMed Abstract | PubMed Central Full Text

   Return to text

6. Yue G, Stratil A, Cepica S, Schroffel J, Schroffelova D, Fontanesi L, Cagnazzo M, Moser G, Bartenschlager H, Reiner G, et al.: Linkage and QTL mapping for Sus scrofa chromosome 7.

   Journal of Animal Breeding and Genetics 2003, 120(s1):56-65. Publisher Full Text

   Return to text

7. de Koning DJ, Rattink AP, Harlizius B, Groenen MAM, Brascamp EW, van Arendonk JAM: Detection and characterization of quantitative trait loci for growth and reproduction traits in pigs.

   Livestock Production Science 2001, 72(3):185-198. Publisher Full Text

   Return to text

8. Tanaka M, Suzuki K, Morozumi T, Kobayashi E, Matsumoto T, Domukai M, Eguchi-Ogawa T, Shinkai H, Awata T, Uenishi H: Genomic structure and gene order of swine chromosome 7q1.1-->q1.2.

   Anim Genet 2006, 37(1):10-16. PubMed Abstract | Publisher Full Text

   Return to text

9. Barbosa A, Demeure O, Urien Cl, Milan D, Chardon P, Renard C: A physical map of large segments of pig Chromosome 7q11-q14: comparative analysis with human Chromosome 6p21.

   Mammalian Genome 2004, 15(12):982-995. PubMed Abstract | Publisher Full Text

   Return to text

10. Tanaka T, Yamamoto J, Iwasaki S, Asaba H, Hamura H, Ikeda Y, Watanabe M, Magoori K, Ioka RX, Tachibana K, et al.: Activation of peroxisome proliferator-activated receptor {delta} induces fatty acid {beta}-oxidation in skeletal muscle and attenuates metabolic syndrome.

    PNAS 2003, 100(26):15924-15929. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

11. Kuehn LA, Rohrer GA, Nonneman DJ, Thallman RM, Leymaster KA: Detection of single nucleotide polymorphisms associated with ultrasonic backfat depth in a segregating Meishan × White Composite population.

    J Anim Sci 2007, 85(5):1111-1119. PubMed Abstract | Publisher Full Text

    Return to text

12. Demars J, Riquet J, Sanchez MP, Billon Y, Hocquette JF, Lebret B, Iannuccelli N, Bidanel JP, Milan D, Gondret F: Metabolic and histochemical characteristics of fat and muscle tissues in homozygous or heterozygous pigs for the body composition QTL located on chromosome 7.

    Physiol Genomics 2007, 30(3):232-241. PubMed Abstract | Publisher Full Text

    Return to text

13. Ponsuksili S, Murani E, Walz C, Schwerin M, Wimmers K: Pre- and postnatal hepatic gene expression profiles of two pig breeds differing in body composition: insight into pathways of metabolic regulation.

    Physiol Genomics 2007, 29(3):267-279. PubMed Abstract | Publisher Full Text

    Return to text

14. Chamary JV, Parmley JL, Hurst LD: Hearing silence: non-neutral evolution at synonymous sites in mammals.

    Nat Rev Genet 2006, 7(2):98. PubMed Abstract | Publisher Full Text

    Return to text

15. Kimchi-Sarfaty C, Oh JM, Kim I-W, Sauna ZE, Calcagno AM, Ambudkar SV, Gottesman MM: A "Silent" Polymorphism in the MDR1 Gene Changes Substrate Specificity.

    Science 2007, 315(5811):525-528. PubMed Abstract | Publisher Full Text

    Return to text

16. Wray GA: The evolutionary significance of cis-regulatory mutations.

    Nat Rev Genet 2007, 8(3):206. PubMed Abstract | Publisher Full Text

    Return to text

17. Luquet S, Lopez-Soriano J, Holst D, Fredenrich A, Melki J, Rassoulzadegan M, Grimaldi PA: Peroxisome proliferator-activated receptor {delta} controls muscle development and oxidative capability.

    FASEB J 2003, 17(15):2299-2301. PubMed Abstract | Publisher Full Text

    Return to text

18. Skogsberg J, Kannisto K, Roshani L, Gagne E, Hamsten A, Larsson C, Ehrenborg E: Characterization of the human peroxisome proliferator activated receptor delta gene and its expression.

    Int J Mol Med 2000, 6(1):73-81. PubMed Abstract

    Return to text

19. Richtlinie für die Stationsprüfung auf Mastleistung, Schlachtkörperwert und Fleischbeschaffenheit beim Schwein [http://www.zds-bonn.de/download.php/790/rlnkp04.pdf] webcite

20. Porcine BAC End Sequencing Project [http://www.sanger.ac.uk/Projects/S_scrofa/BES.shtml] webcite

21. Staden R: The Staden sequence analysis package.

    Mol Biotechnol 1996, 5(3):233-241. PubMed Abstract | Publisher Full Text

    Return to text

22. Primer3 [http://frodo.wi.mit.edu/primer3/] webcite

23. National Center for Biotechnology Information [http://www.ncbi.nlm.nih.gov/] webcite

24. Nickerson DA, Tobe VO, Taylor SL: PolyPhred: automating the detection and genotyping of single nucleotide substitutions using fluorescence-based resequencing.

    Nucleic Acids Res 1997, 25(14):2745-2751. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

25. Ewing B, Hillier L, Wendl MC, Green P: Base-calling of automated sequencer traces using phred. I. Accuracy assessment.

    Genome Res 1998, 8(3):175-185. PubMed Abstract | Publisher Full Text

    Return to text

26. Ewing B, Green P: Base-calling of automated sequencer traces using phred. II. Error probabilities.

    Genome Res 1998, 8(3):186-194. PubMed Abstract | Publisher Full Text

    Return to text

27. Gordon D, Abajian C, Green P: Consed: a graphical tool for sequence finishing.

    Genome Res 1998, 8(3):195-202. PubMed Abstract | Publisher Full Text

    Return to text

28. Frith MC, Hansen U, Weng Z: Detection of cis-element clusters in higher eukaryotic DNA.

    Bioinformatics 2001, 17(10):878-889. PubMed Abstract | Publisher Full Text

    Return to text

29. Gene Regulation [http://www.gene-regulation.com] webcite

30. Cartharius K, Frech K, Grote K, Klocke B, Haltmeier M, Klingenhoff A, Frisch M, Bayerlein M, Werner T: MatInspector and beyond: promoter analysis based on transcription factor binding sites.

    Bioinformatics 2005, 21(13):2933-2942. PubMed Abstract | Publisher Full Text

    Return to text

31. Zuker M: Mfold web server for nucleic acid folding and hybridization prediction.

    Nucl Acids Res 2003, 31(13):3406-3415. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

32. Stephens M, Donnelly P: A comparison of bayesian methods for haplotype reconstruction from population genotype data.

    Am J Hum Genet 2003, 73(5):1162-1169. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

33. Stephens M, Smith NJ, Donnelly P: A new statistical method for haplotype reconstruction from population data.

    Am J Hum Genet 2001, 68(4):978-989. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

34. R: A language and environment for statistical computing [http://www.r-project.org] webcite

35. effects: Effect Displays for Linear and Generalized Linear Models [http://www.r-project.org] webcite

36. Alvarez MJ, Vila-Ortiz GJ, Salibe MC, Podhajcer OL, Pitossi FJ: Model based analysis of real-time PCR data from DNA binding dye protocols.

    BMC Bioinformatics 2007, 8:85. PubMed Abstract | BioMed Central Full Text | PubMed Central Full Text

    Return to text

37. Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR.

    Nucl Acids Res 2001, 29(9):e45. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

        

Abstract

Background

The Glued gene of Drosophila melanogaster encodes the homologue of the vertebrate p150^Glued subunit of dynactin. The Glued¹ mutation compromises the dynein-dynactin retrograde motor complex and causes disruptions to the adult eye and the CNS, including sensory neurons and the formation of the giant fiber system neural circuit.

Results

We performed a 2-stage genetic screen to identify mutations that modified phenotypes caused by over-expression of a dominant-negative Glued protein. We screened over 34,000 flies and isolated 41 mutations that enhanced or suppressed an eye phenotype. Of these, 12 were assayed for interactions in the giant fiber system by which they altered a giant fiber morphological phenotype and/or altered synaptic function between the giant fiber and the tergotrochanteral muscle motorneuron. Six showed interactions including a new allele of atypical protein kinase C (aPKC). We show that this cell polarity regulator interacts with Glued during central synapse formation. We have mapped the five other interacting mutations to discrete chromosomal regions.

Conclusion

Our results show that an efficient way to screen for genes involved in central synapse formation is to use a two-step strategy in which a screen for altered eye morphology precedes the analysis of central synaptogenesis. This has highlighted a role for aPKC in the formation of an identified central synapse.

Background

During the development of a neural connection the axon of the growing neuron has to make morphogenic changes to form the presynaptic apparatus needed for efficient synaptic function once it has reached its target cell. This process involves reception of signals by the presynaptic cell followed by precise rearrangements of the cytoskeleton to direct changes in cell shape and control the formation of the presynaptic apparatus (see [1,2] for reviews).

The giant fiber system (GFS) is a unique neural circuit that contains several of the few identified central synapses in Drosophila and includes the largest in the fly between the giant fiber (GF) interneuron and the leg extensor muscle motorneuron, the tergotrochanteral motorneuron (TTMn), the GF-TTMn synapse [3]. Several studies using over-expression of dominant-negative transgenes, or homozygous adult viable mutations, have recently shed light on signaling mechanisms during the formation of the GF-TTMn synapse. These include the receptors Semaphorin 1a and Roundabout [4,5]; the L-1 type cell-adhesion molecule Neuroglian [6]; the endocytotic and ubiquitin machinery [7-10]; the small GTPase DRac1 [11], and the transcription factor Ken [12]. However, the precise mechanisms by which these integrate during synaptogenesis are yet to emerge.

Glued encodes the largest subunit of the retrograde motor dynein-activating complex dynactin [13,14]. The Glued¹(Gl¹) mutation results in a truncated protein product [15] that disrupts the dynein-dynactin complex by binding to dynein and microtubules but fails to bind to cargoes [16]. Mutants have both visual and CNS defects [17-19]. Glued has a key role in formation of the GF presynaptic bend which may involve local cytoskeletal dynamics and rearrangements [20]. Forward genetic screens to identify gene products involved in post-mitotic neural differentiation can be problematic as many of these genes are plieotropic and will have vital functions earlier in development, thus preventing mutants from being recovered. Moreover, many of the genes may well have several functions during differentiation of a single neuronal type. Consequently, mutants, therefore, will exhibit phenotypes that are difficult to interpret. Indeed, this has been shown for the dynein-dynactin complex in the formation of mushroom body neurons [21]. The eye is not required for either viability or fertility of the adult and genetic disruptions targeted to the eye have been exceptionally useful in deducing signaling pathways, for example the sevenless pathway [22], and also in identifying mutant alleles that would cause lethality earlier in development if they were to be expressed throughout the organism [23-25]. This coupled with the fact that it has been estimated that two thirds of the vital genes within the Drosophila melanogaster genome (~2,500) are involved in its development [26], makes the eye invaluable in a primary screen for genes with additional roles in processes such as neural differentiation.

In this study we have undertaken a genetic modifier screen in the adult eye and isolated 12 mutations that either dominantly enhance or suppress a phenotype caused by over-expression of a dominant-negative form of the Glued protein (Gl^DN). We assayed these mutations for additional interactions with Glued in the GFS and found that 6 show interactions. Mapping of each modifier mutation is presented. One of the suppressors is an allele of aPKC, and we found that other aPKC mutant alleles exhibit suppression of the synaptic phenotype caused by over-expression of Gl^DN.

Results

Screening for modifiers of a truncated Glued (Gl^DN) over-expression eye phenotype

Gl¹ is a true dominant-negative mutation and the effects of the truncated product produced are dose-sensitive [17,20,27]. We previously exploited this by generating a UAS-Gl^DN transgene to over-express this "poison subunit" using the GAL4-UAS system [28] and showed that disruption of retrograde motor function resulted in synaptic defects in the GFS [20]. Our aim was to identify genes that acted with Glued during GFS formation in the developing CNS. Direct screening for alterations of an adult neural phenotype is problematic and extremely labor-intensive because it would involve an F2 screen. Individual mutant stocks would need to be made and crossed into the appropriate mutant background followed by dissection and staining of the adult CNS.

We therefore reasoned that since Gl¹ affects the development of the eye, monitoring the eye phenotype in a primary F1 screen would provide an excellent read-out for defining interacting loci. Using the eye-specific GMR-GAL4 line to target truncated Glued (Gl^DN) resulted in the generation of adult flies possessing smaller eyes with fused ommatidia and miss-arrayed eye bristles. As expected, this was a more severe phenotype than that seen in Gl¹ mutants (Figure 1C &1D). This disruption provided a sensitized background in which to base an F1 screen on adult eyes to isolate novel mutations that altered this phenotype. We performed an EMS screen for second-site modifiers that dominantly enhanced or suppressed the over-expression phenotype in the eye (Figure 1A; see materials and methods). We screened over 34,000 flies and obtained both suppressors (Figure 1E) and enhancers (Figure 1F) that reproducibly altered the eye phenotype. We recovered nine lines with homozygous lethal mutations on the second chromosome and seven lines with homozygous lethal mutations on the third chromosome that were either enhancers of Glued (EGs) or suppressors of Glued (SGs). The lethal mutations were most likely the dominant modifiers of the eye phenotype but we could not rule out the posibility of second-site mutations on the chromosomes that were responsible for altering the phenotype. The lines were crossed inter se for complementation. All mutations showed complementation illustrating each of the 16 mutations map to independent loci (data not shown). Twelve of these were analyzed further (see below).

Identification of mutations that interact with Gl¹ in the adult eye

Our sensitized screen allowed us to isolate mutations that reproducibly dominantly enhanced or suppressed the eye phenotype caused by over-expression of Gl^DN. From these lines we wished to identify which mutations dominantly interacted with Glued, rather than simply up- or down-regulated the GAL4/UAS system or interacted with the synthetic GMR element which can cause disruptions if homozygous or if the temperature is raised [29]. To do this we tested whether these mutations could dominantly interact with the Gl¹ allele by making flies heterozygous for both Gl¹ and the novel EG or SG mutations and identifying subsequent alterations of the Gl¹ eye phenotype. The Gl¹ eye phenotype was clearly exacerbated by the enhancer EG162 (Figure 1C &1G). Other mutations caused more subtle alterations of the Gl¹ adult eye phenotype that were either variable or could not be distinguished when viewed under a dissecting microscope. The eyes of Gl¹/+ individuals show irregular ommatidia and bristle orientation as revealed by SEM (compare Figure 2B to 2A). In addition retinal sections reveal a variable number of rhabdomeres in each ommatidium, which are often reduced in size, as well as clear disruption of the accessory cells. As previously reported, these affects produce a distortion in the overall shape of the ommatidia (Figure 2F), [17]. To determine whether we had isolated interacting mutations we examined whole eyes by SEM and retinal sections by light microscopy from the eyes of Gl¹/+ flies and those transheterozygous with an enhancer (EG37) and a suppressor (SG13). When EG37 was introduced into the Gl¹ background (EG37/+; Gl¹/+) we saw an increase in bristles and an increase in ommatidial fusion and disorganization (Figure 2C). Retinal sections of the eyes from these flies revealed increased disruptions of accessory cells and fused rhabdomeres (Figure 2G). The number of rabdomeres per ommatidium was 4.43 ± 1.11 (n = 195, P < 0.001) compared to those from flies containing Gl¹/+ alone which had 5.03+1.28 (n = 119). We also sometimes saw holes between some ommatidia in the sections (data not shown). With SG13 in the Gl¹ background (SG13/+; Gl¹/+), the ommatidia and bristles became more ordered than is seen in Gl¹ alone (Figure 2, compare D [inset] with B [inset]) and the sections revealed a return to a more hexagonal lattice of accessory cells and trapezoidal pattern of the rhabdomeres (Figure 2H). These contained an average of 5.89 ± 1.13 (n = 195, P < 0.001) rhabdomeres per ommatidium. These data indicated that, for at least two of our mutations, we had isolated loci that genetically interact with Glued during eye development. A third suppressor, SG46, was also analysed in this way and also showed supression of the Gl¹ eye phenotype (data not shown).

Mutations that interact with Glued in the Giant Fiber System and alter axon morphology

The presynaptic terminal of the GF-TTMn synapse is a distinctive bend at the end of the GF axon closely apposed to the TTMn dendrite [30]. Several studies have reported that this bend is often absent or altered when the two neurons fail to form a proper synapse [4,6,7,9,11,20,31]. Since our main aim was to identify genes involved in synapse formation within the GFS we next performed morphological analysis of the giant fiber neurons in adult flies carrying an EG or SG mutation and with disrupted Glued function.

We used the GF-specific GAL4 enhancer-trap line, A307 [GAL4] (hereafter referred to as A307), to identify alterations of the GF morphological phenotype, brought about by Gl^DN over-expression, in the presence of the dominant enhancers or suppressors (Figure 3, Table 1). Our previous work indicated that there are two aspects to the phenotype observed when Gl^DN is targeted to the GF. First, the distinctive terminal bend does not form after 48 hrs of pupal development indicating that synaptogenesis with the TTMn is defective. Second, as the GF axon develops during later pupal stages, the tip swells, often to several times the axon diameter, presumably because of a build-up of cellular material that the motor is unable to move toward the cell body [20]. Like many GAL4-generated phenotypes there is some variation between preparations, most notably on the extent of axon swelling (compare Figure 3C and 6B). While both the lack of terminal bend and axon swelling are a result of retrograde motor disruption, any link between the two phenotypes is unclear. Both SG13/+ and SG46/+ showed suppression of these two phenotypes when Gl^DN was expressed using A307 (Figure 3F &3H, Table 1). Swollen axon tips were rarely seen and the axons were either "bendless," or showed at least one bend in a preparation (Figure 3F [arrowhead] &3H, Table 1), a phenotype not seen when the dominant-negative subunit is expressed alone (Table 1). SG58/+ showed a mild suppression of the GF morphology phenotype with consistently fewer neurons exhibiting swollen axons (Table 1, see below). However no bends or partial bends were observed (Figure 3G). We could not unequivocally rule out the possibility of these mutations altering the effectiveness of the GAL4/UAS system, however, the phenotypes observed for SG13/+ and SG46/+ were different from those seen if GAL4 expression was suppressed by reducing temperature (M.J.A unpublished observations). SG16/+ showed no rescue of the GF morphology phenotype, suggesting that this mutation does not interact with Glued in the GFS. The EG162/+ and EG165/+ caused dominant lethality in the A307>Gl^DN background (Table 1) presumably because they affected other cells expressing Gl^DN in the A307 line that eliminated viability. 20% (4/20) of EG28/+ preparations exhibited GF axons which remained in the brain (Figure 3E). This is likely to be due to a failure to exit the brain on outgrowth, or retraction after a failure in synatogenesis. No obvious enhancement of the GF phenotype was observed with EG37/+ (Figure 3D).

We reasoned that mutations in genes involved in synaptogenesis may not greatly enhance the already severe Gl^DN synaptic phenotype. Mutations in genes that have a role earlier in development, for example in axon guidance, may give rise to a discernable enhancement, such as incorrect GF axon growth as was the case for EG28/+. In Gl¹ mutants the GFs look morphologically normal, although they have GF-TTMn synaptic defects [20]. Therefore, we tested for dominant effects of the EG mutations on the morphology of the GFs in a Gl¹/+ background by generating flies with A307, UAS-LacZ, Gl¹ and our third chromosome EGs (Figure 3I-K, Table 1). We saw an increase in the frequency of "bendless" axons in 3 of 4 EG mutations tested, suggesting an enhancement of the Gl¹/+ phenotype, but no appearance of the swellings seen with over-expression of the dominant-negative subunit. Unusually, preparations containing EG28/+ and Gl¹/+ exhibited a reduction in "bendless" axons and often had ectopic axonal branching (Figure 3K). Taken together with the phenotype seen with Gl^DN (see above), we presume this mutation is at a locus that has a role in axon growth and/or guidance. We were unable to do this with the mutations on chromosome two, using any GAL4 GF marker lines, since the mutations are on the same chromosome as the UAS-Gl^DN insert resulting in over-expression of Gl^DN in any cell that express GAL4 and contain the mutation. The UAS-Gl^DN element and the interacting mutations would need to be separated onto different chromosomes by either recombination or precise excision of the P-element containing the UAS-Gl^DN transgene. As an alternative, morphology of the GF can be observed using dye-filling techniques [31-33]. However, these techniques are labor-intensive and were not practicable for screening purposes.

Mutations that interact with Glued in the Giant Fiber System and alter synaptic function

We used electrophysiology to assay GFS function, which, in combination with morphological analysis, can reveal abnormalities of synaptogenesis during development [4-7,9,11,12,20]. A severe GF-TTMn synaptic phenotype is caused by over-expression of Gl^DN ; some flies fail to respond to brain stimulation and those that do respond exhibit a long response latency and show only a single response or poor following to repetitive stimuli ([20]; Figure 4). We put the SG mutations which had exhibited morphological rescue into the A307/UAS-Gl^DN background. Corresponding with the observed morphological suppression (Figure 3F &3H, Table 1), flies also containing either SG13/+ or SG46/+ exhibited an increase in GF-TTMn synaptic function. All preparations responded to brain stimulation and had shorter response latencies and increased following at 100 and 250 Hz than is seen with simple over-expression of Gl^DN (Figure 4). SG58/+ also exhibited a detectable increase in synaptic function with all preparations responding upon stimulation, a slight reduction in the long latency seen when Gl^DN is over-expressed and also a corresponding increase in following to repetitive stimuli which was significant at 100 Hz (Figure 4).

Gl¹/+ mutants show a functional defect in the GF-TTMn synapse when tested using electrophysiology. They exhibit a response latency not significantly different from wild type (χ = 1.1 ms, n = 8) to brain stimulation, but do not follow 1:1 on stimulation at a frequency of 250 Hz as observed in wild type flies ([20]; Figure 5A &5C). We crossed the EGs into the Gl¹/+ background to observe any enhancement of this synaptic phenotype. Because this test did not rely on GAL4 expression we were able to assay any of the EG mutations, regardless of their chromosomal location. All of the EGs tested showed significantly increased reponse latencies (Figure 5B) and exhibited poor following. To analyse this more carefully we looked at the probability of a response being seen with each sequential stimulus at 250 Hz. When 10 stimuli were given to Gl¹/+ mutants, the flies demonstrated a depression in response to stimuli 1 through 10 with a plateau after stimulus 6 (Figure 5D), probably due to TTMn not always reaching threshold. This resulted in a 30% (probability 0.3) chance of responding to the last 4 stimuli for Gl¹/+ flies whereas wild type flies will typically have > 90% (probability > 0.9) chance of responding (Figure 5D). When EG37/+ or EG162/+ were trans-heterozygous with Gl¹/+ an enhancement of the phenotype was seen since 100% of Gl¹/+ flies responded to stimulus number 3 whereas the probability of responding fell to 0.6 with EG37/+ and <0.2 with EG162/+ added (Figure 5D). We used root mean square deviation (RMSD) to measure the differences in the depression curves and compared them to that seen with Gl¹/+. The average deviation between Gl¹/+; EG37/+ and Gl¹/+ was 0.23 (P < 0.1) and between Gl¹/+; EG162/+ and Gl¹/+ was 0.48 (P < 0.001). For most of the other enhancers tested there was no significant enhancement of the depression seen with Gl¹/+ alone (Figure 5E). This included EG28/+ which had given outgrowth/retraction phenotypes when combined with Gl^DN (see Figure 3). Two of the EGs, EG56/+ & EG79/+, seemed to suppress the depression (Figure 5F) despite the fact that they caused an increase in the reponse latency (Figure 5B). The reason for this is not known.

aPKC, but not Su(H), interacts with Glued in synapse formation

We mapped the six mutations that showed significant interactions with Glued in the GFS using deficiencies and known lethal alleles (see materials and methods). This was assuming that the lethality was associated with the enhancer or supressor. A caveat being that lethality could be due to a second mutation and the interaction with Glued due to a viable allele. Two of the mutations mapped to known genes, aPKC and Su(H), and the others were mapped to small regions on chromosome two or three (Table 2). To determine whether the suppression of the Gl^DN phenotype in the GF by SG58 was due to the mutation in aPKC we recombined both the aPKC⁰⁶⁴⁰³ and aPKC^EY22496 alleles onto the UAS-Gl^DN chromosome and crossed these lines to A307. Morphological examination of the GF axons revealed that they were less swollen than seen in A307>UAS-Gl^DN preparations and occasionally were seen to have a bend (Figure 6). GF-TTMn synaptic function was also increased compared to A307>Gl^DN flies. When tested using electrophysiology, 89% (16/18) of A307>Gl^DN, aPKC⁰⁶⁴⁰³/+ flies responded upon GF stimulation compared to 66% (10/15) of A307>Gl^DN flies and they exhibited a slightly reduced latency (Figure 6F &6G) and a statistically significant increase in following at 100 Hz (Figure 6F &6H). With the aPKC^EY22496 allele, weaker suppression was seen with 78% (14/18) responding and they exhibited an increase in following at 100 Hz (Figure 6F &6H). Overall, the results show weak support for the aPKC interaction with A307>UAS-Gl^DN from the aPKC⁰⁶⁴⁰³ allele and weaker support from the aPKC^EY22496 allele, however, they are very similar to the effect of the SG58/+ (Figures 3 &4) and indicate that aPKC interacts with Glued genetically to alleviate the phenotype caused by the over-expression of the poison subunit.

To determine whether the enhancement of the Gl¹/+ phenotype in the GF by EG37/+ was due to the mutation in Su(H) we generated Su(H)¹/+; Gl¹/+ double heterozygotes and again looked at both the morphology of the GFs, with our A307 line, and performed electrophysiology to assay the function of the GF-TTMn synapse. No enhancement of the Gl¹/+ phenotype was observed in these flies (data not shown). We cannot therefore rule out the presence of a second unmapped mutation on the EG37 chromosome. Alternatively, the mutation in EG37 may have different properties than the Su(H)¹ allele.

Discussion

The success of our two-stage screening approach may have been facilitated by the fact that Glued has a plethora of distinct roles during eye development, including organizing optic neural architecture [17,34,35] and an involvement in the formation of sensory neuronal circuits [18,19]. Therefore we had an eye phenotype on which to base the screen. However, this does not preclude such a method being used for identifying genes involved in other aspects of neural differentiation. We found that 50% (6/12) of the isolated mutation-containing chromosomes that altered the eye phenotype also altered GFS phenotypes when tested.

The over-expression of the truncated Glued protein caused strong phenotypes in both the eye and GF neurons, greater than those caused by heterozygosity for the dominant Gl¹ allele. This is likely to be due to the GAL4-UAS system producing many more molecules of the truncated product than Gl¹/+ cells in which, theoretically, a maximum of half of the Glued molecules will be truncated. Consistent with this observation, both the suppressors and enhancers isolated during this screen showed stronger effects on Gl^DN eye phenotypes than on those produced by Gl¹. Determining interactions with the Gl¹allele also allowed us to confirm GAL4-independent interactions with the Glued locus. For all of the mutations (with the exception of EG162), the alterations of the weaker Gl¹/+ eye phenotype were not obvious, however, SEM and sectioning was performed to show interactions with two of the mutations (EG37 & SG13).

We used two different disruptions of Glued function, one strong and the other weaker, to assay successfully the effects of both enhancer and suppressor mutations in the GFS using both morphological and electrophysiological criteria. The severe disruptions of GF morphology and synaptic function enabled the effects of suppressor mutations to be clearly observed. This was less reliable when assaying the effects of mutations isolated as enhancers as either no increase of the already severe phenotype was seen or the interaction was lethal. For the enhancers, therefore, we relied on generating double heterozygotes with Gl¹/+. As was the case in the eye, interactions were less pronounced and only two enhancers, EG37/+ and EG162/+ showed enhancement of the Gl¹/+ electrophysiological phenotype. Indeed, the subtlety of some interactions with Gl¹/+ may have resulted in our analyses being unable to detect some positive interacting loci in the GFS that altered the eye phenotype caused by Gl^DN.

We have generated some EMS alleles, two of which we have mapped to known genetic loci (see below) and four of which we have mapped to discrete chromosomal locations (Table 2). However, these four complement all the available lethal alleles in these regions indicating that our mutations lie in loci for which there are few or no lethal alleles available. Identification of the location of these new alleles will require either new rounds of mutagenesis, such as via P-element excision in the mapped regions, finer mapping using SNPs [36-38] or custom made deficiencies using stocks from the DrosDel project [39]. Completion of the BDGP Gene Disruption Project may also enable mapping of the lesions [40] along with more recent approaches using other transposable elements that may disrupt genes refractory to P-element disruption [41-43]. Interestingly, we appear not to have isolated any mutations in genes that encode known components of the retrograde motor complex including any further alleles of Glued. During some of the early genetic analysis of the Glued locus, dominant second-site suppressors of the Gl¹ eye phenotype were isolated and reported [27]. Of these, two were mapped to the X chromosome (Su [Gl]27 &Su [Gl]57,[27]) and the others, Su(Gl)77 &Su(Gl)102 are alleles of Dynein heavy chain 64C [44,45]. From the map positions of our mutations, we have not re-isolated similar alleles. Because our primary screen involved making only the eye mutant for Glued, we could potentially isolate mutations that are lethal in combination with Gl¹ and would, therefore, not have been isolated Harte and Kankel's screen. However, none of the enhancers tested were lethal with Gl¹.

We have successfully isolated two new alleles of known genes, Su(H) and aPKC. Of the two, we have shown that alleles of aPKC genetically interact with Glued in the GFS and suppress the abnormalities in GF-TTMn synapse formation seen when the retrograde motor complex is compromised by Gl^DN. These abnormalities are: a lack of the presynaptic "bends"; a branching event that takes place after the two neurons have met [31]; swollen axon tips and a weak or absent functional synapse [20]. aPKC is part of a protein complex, with PAR-3 (Bazooka in Drosophila) and PAR6 that regulates cell polarity in a number of different tissues/cells of Drosophila and vertebrates including neurons [46-48]. So what is the role of aPKC in the GF neuron? In vertebrate neurons aPKC is needed for neurite outgrowth [49-51]. In contrast, aPKC in flies is an essential part of the machinery that polarizes dividing neuroblasts [52] but is not needed postmitotically for outgrowth [53]. Our data also indicate that aPKC is not needed for neurite extension since the introduction of aPKC mutations into our sensitized background has no effect on GF outgrowth. aPKC is involved in memory formation in Drosophila [54] and at the developing larval NMJ it regulates microtubules (MTs) both pre- and postsynaptically during synapse formation [55]. Indeed MTs are one of the major targets of the PAR-3/PAR-6/aPKC complex in several contexts [56-58]. aPKC regulates MT orientation in fibroblasts [59,60] and MT organization in the early embryo [61]. At the NMJ it controls MT stability with a reduction in aPKC activity causing a decreased association of MTs with the microtubule associated protein Futsch and MT fragmentation [55]. Dynein-dynactin is known to be involved in MT organization during growth cone remodeling as well as polarizing MTs in axons [62,63]. Our data indicate that dynein-dynactin and aPKC are acting antagonistically during formation of the GF presynaptic structure and suggest that both are needed to control microtubule organization and dynamics in synapse formation but have opposing roles. One simple explanation is that one of the roles of dynein-dynactin in the GF is to alter MT dynamics at the tip of the axon, when it has reached its post-synaptic target, so that they are more mobile enabling the presynaptic bend to be formed. aPKC regulates the stability of MTs thereby confining axon branching to a single bend. Blocking dynein-dynactin function prevents the MT re-organization needed for formation of the bends and this is ameliorated when aPKC function is reduced.

Conclusion

We have used a novel approach to screen for genes involved in central synapse formation by performing a primary screen, using a sensitized background, on the adult eye and then a secondary screen, on the isolated mutations, for synaptic phenotypes. This study shows that forward genetic screens are powerful tools for identifying genes with roles in CNS development.

Methods

Drosophila strains

The A307 [GAL4] enhancer-trap line and the line containing the UAS-Gl^DN transgene on chromosome 2 (UAS-Gl^Δ96B) have been described previously [20,33,64], as has the eye specific GMR-GAL4 line [65]. Stocks containing balancers, deficiency stocks, lethal alleles, aPKC⁰⁶⁴⁰³, aPKC^EY22496 and Su(H)¹ flies were obtained from the Bloomington Drosophila Stock Center, Indiana, USA.

Gl^DN over-expression screen

w; UAS-Gl^DN males, isogenized on the second and third chromosomes, were mutagenized by feeding overnight with a 0.25 mM solution of EMS in 1% sucrose. This dose was to generate, on average, only one lethal mutation per genome to facilitate downstream analysis. The mutagenized males were then mated to w; GMR-GAL4; TM6B/MKRS virgin females and the eyes of the progeny were scored for an enhancement or suppression of the Gl^DN eye phenotype. We isolated 324 adults with altered eye morphology; 215 potential enhancers and 109 potential suppressors. Mutations were recovered by mating flies with altered eye phenotypes individually to w; CyO/Sco; MKRS/TM6B flies. Sibling crosses were then performed to obtain lines with recessive lethal, or recessive viable, mutations on chromosomes 2 or 3 balanced over CyO, MKRS or TM6B. One hundred and sixteen lines were recovered with either a recessive lethal on chromosome 2, 3, or both. We did not obtain any recessive mutations that gave a homozygous visible phenotype. The 116 lines were crossed back to w; GMR-GAL4; TM6B/MKRS flies to confirm that the recovered mutations altered the GMR-GAL4/UAS-Gl^DN eye phenotype. This resulted in 26 lines containing mutations that enhanced the phenotype and 15 lines that suppressed the phenotype.

Scanning electron microscopy (SEM) and Retinal sectioning

Whole flies were fixed in 2% glutaraldehyde and dehydrated in acetone. Samples were dried in a Polatron E3000 critical point dryer, mounted onto stubs, and coated with gold. Micrographs were taken on a Hitachi S-430 electron microscope. For retinal sections, head cases were dissected from whole flies and fixed in 2% glutaraldehyde, 2% paraformaldehyde in 0.1 M PBS (0.1 M NaCl, 0.1 M Na₂HPO₄/NaH₂PO₄ [pH 6.8]) overnight at 4°C. Following dehydration in an acetone series, the head cases were embedded in Durcupan resin and 2.5 μm sections cut with a Lieca: Jung RM2065 microtome. Sections were stained with Toluidine blue and photographed on a Leica DMR microscope using a Leica DC500 digital camera.

CNS Histochemistry

The central nervous systems were dissected from adult flies in 0.1 M PBS with 0.05% Triton X-100. For X-Gal staining they were then fixed in 1% gluteraldehyde for 5 min. Preparations were washed in PBT (0.1 M PBS, 0.1% Triton X-100) and pre-warmed in 2 mls of X-Gal staining solution (3 mM K₄ [Fe(CN)₆], 3 mM K₃ [Fe(CN)₆], 1 mM MgCl₂, 150 mM NaCl, 10 mM Na₂HPO₄/NaH₂PO₄ [pH 7.2], 0.3% Triton X-100) in a watch glass for 5 min at 37°C. To this was added 1 ml of staining solution, saturated with dissolved X-Gal (5-bromo-4-chloro-3-indolyl β-D-galactopyranoside) and pre-warmed to 37°C, and the preparations were incubated for several hours until staining of the GFs was complete.

Electrophysiology

Recordings from the GFS of adult flies were made essentially as described in [20]; a method based on those described by [66] and [67]. Flies were cooled on ice until they were immobile and secured in wax, ventral side down, with the wings held outwards in the wax. The GFs were stimulated using a Grass S48 stimulator to deliver a 40 V pulse for 0.03 ms through tungsten electrodes pushed through the eyes and into the brain. A tungsten earth wire was placed into the abdomen. Glass microelectrodes (resistance 40-60 MΩ), filled with 3 M KCl, were driven through the cuticle into the TTM and DLM muscles to record responses. These were amplified using Getting 5A amplifiers (Getting Instruments, USA) and the data digitized using an analogue-digital Digidata 1320 and Axoscope 9.0 software (Axon Instruments, USA). A single pulse was delivered for response latency measurements and trains of 10 stimuli, either 4 ms (250 Hz) or 10 ms (100 Hz) apart, were given with a 5 s interval between each train for following frequency recordings. We routinely recorded from both TTM and a DLM in each preparation to ensure that correct stimulation of the GF was achieved. However, only data from the TTM are presented in this report.

Genetic mapping

Mutations were mapped using the "classic"second chromosome "deficiency kit" (DK2) of 110 stocks from the Bloomington Drosophila Stock Center containing mostly deletions that were defined cytologically. Once deficiencies were identified that failed to complement the isolated lethal mutations, smaller deficiencies in the regions were tested for complementation including molecularly defined deficiencies recently made available from Exelixis and the DrosDel project [39,68]. Standard complementation tests were then performed with known lethal mutations in the chromosomal regions uncovered by the deficiencies. SG13 failed to complement Df(2R)kr10 but complemented Df(2R)gsb and Df(2R)kr14 indicating that SG13 lies at 60F5. SG46 failed to complement Df(2L)79b which covers 22A2 to 22E1. Testing smaller deletions within this region revealed that SG46 failed to complement the molecularly defined deletions Df(2L)Exel17008 (22B8;22D1) and Df(2L)Exel17011 (22E1;22F3). However when crossed together the two deletions failed to complement each other and both fail to complement alleles of Dpp which lies at 22F1-2. Therefore they both delete at least the region 22E1 to 22F3 and Df(2L)Exel17008 is incorrectly annotated. SG58 failed to complement Df(2R)KnSA4 and Df(2R)XTE-58 indicating it resided between 51D1 and 51D6 on chromosome 2. Using the available lethal mutations we determined SG58 to be an allele of Drosophila atypical Protein Kinase C (aPKC) as it failed to complement aPKC⁰⁶⁴⁰³ and aPKC^EY22496, both known null alleles [52]. EG28 failed to complement Df(3R)p13 which extends from 84F1 to 85A8. Within this region EG28 failed to complement Df(3R)exel6417 but complemented Df(3R)dsx29 making its location to be within 84F7-84F12. All available lethal mutations were found to be not allelic to EG28. EG37 failed to complement Df(2L)TE35BC-24 and Df(2L)A48 and testing known lethal alleles uncovered by both these deficiencies revealed the mutation to be in the Su(H) locus. EG37 is lethal when trans-heterozygous with the Su(H)¹ allele. EG162 failed to complement Df(2R)X1 and Df(2R)stan1 indicating that is resides between 46D and 47A. However, it did complement Df(2R)stan2 indicating that it lies at 46D-E. We have been unable to map this mutation further at present.

Authors' contributions

LM and LAJ carried out the genetic screen; LAJ performed the SEM and eye sectioning; LM and MJA did the electrophysiology, all authors performed CNS dissection and staining. MJA designed and coordinated the work and drafted the manuscript. All authors read and approved the final manuscript.

Acknowledgements

Thank you to Kevin Moffat and Nara Muraro for useful discussions and help with the retinal sections, thanks to Ray Newsam and Ian Brown for help with the SEM. Thank you to Dr Dan Mulvihill for critical reading of the manuscript. This work was supported by a BBSRC studentship to LAJ and a Wellcome Trust Project Grant (069710/Z/02/Z) to MJA.

References

1. Davis GW: The making of a synapse: Target-derived signals and presynaptic differentiation.

   Neuron 2000, 26(3):551-554. PubMed Abstract | Publisher Full Text

   Return to text

2. Jin Y, Garner CC: Molecular mechanisms of presynaptic differentiation.

   Annu Rev Cell Dev Biol 2008, 24:237-262. PubMed Abstract | Publisher Full Text

   Return to text

3. Allen MJ, Godenschwege TA, Tanouye MA, Phelan P: Making an escape: development and function of the Drosophila giant fibre system.

   Semin Cell Dev Biol 2006, 17(1):31-41. PubMed Abstract | Publisher Full Text

   Return to text

4. Godenschwege TA, Hu H, Shan-Crofts X, Goodman CS, Murphey RK: Bi-directional signaling by Semaphorin 1a during central synapse formation in Drosophila.

   Nat Neurosci 2002, 5(12):1294-1301. PubMed Abstract | Publisher Full Text

   Return to text

5. Godenschwege TA, Simpson JH, Shan X, Bashaw GJ, Goodman CS, Murphey RK: Ectopic expression in the giant fiber system of Drosophila reveals distinct roles for roundabout (Robo), Robo2, and Robo3 in dendritic guidance and synaptic connectivity.

   J Neurosci 2002, 22(8):3117-3129. PubMed Abstract | Publisher Full Text

   Return to text

6. Godenschwege TA, Kristiansen LV, Uthaman SB, Hortsch M, Murphey RK: A conserved role for Drosophila Neuroglian and human L1-CAM in central-synapse formation.

   Curr Biol 2006, 16(1):12-23. PubMed Abstract | Publisher Full Text

   Return to text

7. Muralidhar MG, Thomas JB: The Drosophila Bendless Gene Encodes a Neural Protein Related to Ubiquitin-Conjugating Enzymes.

   Neuron 1993, 11(2):253-266. PubMed Abstract | Publisher Full Text

   Return to text

8. Murphey RK, Froggett SJ, Caruccio P, Shan-Crofts X, Kitamoto T, Godenschwege TA: Targeted expression of shibire ts and semaphorin 1a reveals critical periods for synapse formation in the giant fiber of Drosophila.

   Development 2003, 130(16):3671-3682. PubMed Abstract | Publisher Full Text

   Return to text

9. Oh CE, McMahon R, Benzer S, Tanouye MA: Bendless, a Drosophila Gene Affecting Neuronal Connectivity, Encodes a Ubiquitin-Conjugating Enzyme Homolog.

   Journal Of Neuroscience 1994, 14(5 Pt 2):3166-3179. PubMed Abstract | Publisher Full Text

   Return to text

10. Uthaman SB, Godenschwege TA, Murphey RK: A mechanism distinct from highwire for the Drosophila ubiquitin conjugase bendless in synaptic growth and maturation.

    J Neurosci 2008, 28(34):8615-8623. PubMed Abstract | Publisher Full Text

    Return to text

11. Allen MJ, Shan X, Murphey RK: A role for Drosophila Drac1 in neurite outgrowth and synaptogenesis in the giant fiber system.

    Mol Cell Neurosci 2000, 16(6):754-765. PubMed Abstract | Publisher Full Text

    Return to text

12. Allen MJ, Drummond JA, Sweetman DJ, Moffat KG: Analysis of two P-element enhancer-trap insertion lines that show expression in the giant fibre neuron of Drosophila melanogaster.

    Genes Brain Behav 2007, 6(4):347-358. PubMed Abstract | Publisher Full Text

    Return to text

13. Schroer TA: Dynactin.

    Annu Rev Cell Dev Biol 2004, 20:759-779. PubMed Abstract | Publisher Full Text

    Return to text

14. Watermanstorer CM, Holzbaur ELF: The Product Of the Drosophila Gene, Glued, Is the Functional Homolog Of the P150(Glued) Component Of the Vertebrate Dynactin Complex.

    Journal Of Biological Chemistry 1996, 271(2):1153-1159. PubMed Abstract | Publisher Full Text

    Return to text

15. Swaroop A, Paco-Larson ML, Garen A: Molecular genetics of a transposon-induced dominant mutation in the Drosophila locus Glued.

    Proc Natl Acad Sci USA 1985, 82(6):1751-1755. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

16. McGrail M, Gepner J, Silvanovich A, Ludmann S, Serr M, Hays TS: Regulation Of Cytoplasmic Dynein Function In-Vivo By the Drosophila Glued Complex.

    Journal Of Cell Biology 1995, 131(2):411-425. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

17. Fan SS, Ready DF: Glued participates in distinct microtubule-based activities in Drosophila eye development.

    Development 1997, 124(8):1497-1507. PubMed Abstract | Publisher Full Text

    Return to text

18. Murphey RK, Caruccio PC, Getzinger M, Westgate PJ, Phillis RW: Dynein-dynactin function and sensory axon growth during Drosophila metamorphosis: A role for retrograde motors.

    Dev Biol 1999, 209(1):86-97. PubMed Abstract | Publisher Full Text

    Return to text

19. Reddy S, Jin P, Trimarchi J, Caruccio P, Phillis R, Murphey RK: Mutant molecular motors disrupt neural circuits in Drosophila.

    J Neurobiol 1997, 33(6):711-723. PubMed Abstract | Publisher Full Text

    Return to text

20. Allen MJ, Shan X, Caruccio P, Froggett SJ, Moffat KG, Murphey RK: Targeted expression of truncated glued disrupts giant fiber synapse formation in Drosophila.

    J Neurosci 1999, 19(21):9374-9384. PubMed Abstract | Publisher Full Text

    Return to text

21. Reuter JE, Nardine TM, Penton A, Billuart P, Scott EK, Usui T, Uemura T, Luo L: A mosaic genetic screen for genes necessary for Drosophila mushroom body neuronal morphogenesis.

    Development 2003, 130(6):1203-1213. PubMed Abstract | Publisher Full Text

    Return to text

22. Karim FD, Chang HC, Therrien M, Wassarman DA, Laverty T, Rubin GM: A screen for genes that function downstream of Ras1 during Drosophila eye development.

    Genetics 1996, 143(1):315-329. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

23. Carrera P, Abrell S, Kerber B, Walldorf U, Preiss A, Hoch M, Jackle H: A modifier screen in the eye reveals control genes for Kruppel activity in the Drosophila embryo.

    Proc Natl Acad Sci USA 1998, 95(18):10779-10784. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

24. Mahoney MB, Parks AL, Ruddy DA, Tiong SY, Esengil H, Phan AC, Philandrinos P, Winter CG, Chatterjee R, Huppert K, et al.: Presenilin-based genetic screens in Drosophila melanogaster identify novel notch pathway modifiers.

    Genetics 2006, 172(4):2309-2324. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

25. Thomas BJ, Wassarman DA: A fly's eye view of biology.

    Trends Genet 1999, 15(5):184-190. PubMed Abstract | Publisher Full Text

    Return to text

26. Thaker HM, Kankel DR: Mosaic analysis gives an estimate of the extent of genomic involvement in the development of the visual system in Drosophila melanogaster.

    Genetics 1992, 131(4):883-894. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

27. Harte PJ, Kankel DR: Genetic-Analysis Of Mutations At the Glued Locus and Interacting Loci In Drosophila-Melanogaster.

    Genetics 1982, 101(3-4):477-501. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

28. Brand AH, Perrimon N: Targeted Gene-Expression As a Means Of Altering Cell Fates and Generating Dominant Phenotypes.

    Development 1993, 118(2):401-415. PubMed Abstract | Publisher Full Text

    Return to text

29. Kramer JM, Staveley BE: GAL4 causes developmental defects and apoptosis when expressed in the developing eye of Drosophila melanogaster.

    Genet Mol Res 2003, 2(1):43-47. PubMed Abstract | Publisher Full Text

    Return to text

30. Blagburn JM, Alexopoulos H, Davies JA, Bacon JP: Null mutation in shaking-B eliminates electrical, but not chemical, synapses in the Drosophila giant fiber system: a structural study.

    J Comp Neurol 1999, 404(4):449-458. PubMed Abstract | Publisher Full Text

    Return to text

31. Jacobs K, Todman MG, Allen MJ, Davies JA, Bacon JP: Synaptogenesis in the giant-fibre system of Drosophila: interaction of the giant fibre and its major motorneuronal target.

    Development 2000, 127(23):5203-5212. PubMed Abstract | Publisher Full Text

    Return to text

32. Koto M, Tanouye MA, Ferrus A, Thomas JB, Wyman RJ: The Morphology Of the Cervical Giant Fiber Neuron Of Drosophila.

    Brain Research 1981, 221(2):213-217. PubMed Abstract | Publisher Full Text

    Return to text

33. Phelan P, Nakagawa M, Wilkin MB, Moffat KG, O'Kane CJ, Davies JA, Bacon JP: Mutations in shaking-B prevent electrical synapse formation in the Drosophila giant fiber system.

    J Neurosci 1996, 16(3):1101-1113. PubMed Abstract | Publisher Full Text

    Return to text

34. Harte PJ, Kankel DR: Analysis of visual system development in Drosophila melanogaster: mutations at the Glued locus.

    Dev Biol 1983, 99(1):88-102. PubMed Abstract | Publisher Full Text

    Return to text

35. Whited JL, Cassell A, Brouillette M, Garrity PA: Dynactin is required to maintain nuclear position within postmitotic Drosophila photoreceptor neurons.

    Development 2004, 131(19):4677-4686. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

36. Hoskins RA, Phan AC, Naeemuddin M, Mapa FA, Ruddy DA, Ryan JJ, Young LM, Wells T, Kopczynski C, Ellis MC: Single nucleotide polymorphism markers for genetic mapping in Drosophila melanogaster.

    Genome Res 2001, 11(6):1100-1113. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

37. Martin SG, Dobi KC, St Johnston D: A rapid method to map mutations in Drosophila.

    Genome Biol 2001, 2(9):RESEARCH0036. PubMed Abstract | BioMed Central Full Text | PubMed Central Full Text

    Return to text

38. Chen D, Berger J, Fellner M, Suzuki T: FLYSNPdb: a high-density SNP database of Drosophila melanogaster.

    Nucleic acids research 2009, (37 Database):D567-570. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

39. Ryder E, Blows F, Ashburner M, Bautista-Llacer R, Coulson D, Drummond J, Webster J, Gubb D, Gunton N, Johnson G, et al.: The DrosDel collection: a set of P-element insertions for generating custom chromosomal aberrations in Drosophila melanogaster.

    Genetics 2004, 167(2):797-813. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

40. Bellen HJ, Levis RW, Liao G, He Y, Carlson JW, Tsang G, Evans-Holm M, Hiesinger PR, Schulze KL, Rubin GM, et al.: The BDGP gene disruption project: single transposon insertions associated with 40% of Drosophila genes.

    Genetics 2004, 167(2):761-781. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

41. Hacker U, Nystedt S, Barmchi MP, Horn C, Wimmer EA: piggyBac-based insertional mutagenesis in the presence of stably integrated P elements in Drosophila.

    Proc Natl Acad Sci USA 2003, 100(13):7720-7725. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

42. Horn C, Offen N, Nystedt S, Hacker U, Wimmer EA: piggyBac-based insertional mutagenesis and enhancer detection as a tool for functional insect genomics.

    Genetics 2003, 163(2):647-661. PubMed Abstract | PubMed Central Full Text

    Return to text

43. Schuldiner O, Berdnik D, Levy JM, Wu JS, Luginbuhl D, Gontang AC, Luo L: piggyBac-based mosaic screen identifies a postmitotic function for cohesin in regulating developmental axon pruning.

    Developmental cell 2008, 14(2):227-238. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

44. Iyadurai SJ, Robinson JT, Ma L, He Y, Mische S, Li MG, Brown W, Guichard A, Bier E, Hays TS: Dynein and Star interact in EGFR signaling and ligand trafficking.

    Journal of cell science 2008, 121(Pt 16):2643-2651. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

45. McGrail M, Gepner J, Silvanovich A, Ludmann S, Serr M, Hays TS: Regulation of cytoplasmic dynein function in vivo by the Drosophila Glued complex.

    J Cell Biol 1995, 131(2):411-425. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

46. Macara IG: Parsing the polarity code.

    Nature reviews 2004, 5(3):220-231. PubMed Abstract | Publisher Full Text

    Return to text

47. Wiggin GR, Fawcett JP, Pawson T: Polarity proteins in axon specification and synaptogenesis.

    Developmental cell 2005, 8(6):803-816. PubMed Abstract | Publisher Full Text

    Return to text

48. Wodarz A: Molecular control of cell polarity and asymmetric cell division in Drosophila neuroblasts.

    Current opinion in cell biology 2005, 17(5):475-481. PubMed Abstract | Publisher Full Text

    Return to text

49. Shi SH, Jan LY, Jan YN: Hippocampal neuronal polarity specified by spatially localized mPar3/mPar6 and PI 3-kinase activity.

    Cell 2003, 112(1):63-75. PubMed Abstract | Publisher Full Text

    Return to text

50. Nishimura T, Kato K, Yamaguchi T, Fukata Y, Ohno S, Kaibuchi K: Role of the PAR-3-KIF3 complex in the establishment of neuronal polarity.

    Nature cell biology 2004, 6(4):328-334. PubMed Abstract | Publisher Full Text

    Return to text

51. Schwamborn JC, Puschel AW: The sequential activity of the GTPases Rap1B and Cdc42 determines neuronal polarity.

    Nat Neurosci 2004, 7(9):923-929. PubMed Abstract | Publisher Full Text

    Return to text

52. Rolls MM, Albertson R, Shih HP, Lee CY, Doe CQ: Drosophila aPKC regulates cell polarity and cell proliferation in neuroblasts and epithelia.

    J Cell Biol 2003, 163(5):1089-1098. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

53. Rolls MM, Doe CQ: Baz, Par-6 and aPKC are not required for axon or dendrite specification in Drosophila.

    Nat Neurosci 2004, 7(12):1293-1295. PubMed Abstract | Publisher Full Text

    Return to text

54. Drier EA, Tello MK, Cowan M, Wu P, Blace N, Sacktor TC, Yin JC: Memory enhancement and formation by atypical PKM activity in Drosophila melanogaster.

    Nat Neurosci 2002, 5(4):316-324. PubMed Abstract | Publisher Full Text

    Return to text

55. Ruiz-Canada C, Ashley J, Moeckel-Cole S, Drier E, Yin J, Budnik V: New synaptic bouton formation is disrupted by misregulation of microtubule stability in aPKC mutants.

    Neuron 2004, 42(4):567-580. PubMed Abstract | Publisher Full Text

    Return to text

56. Etienne-Manneville S, Hall A: Cell polarity: Par6, aPKC and cytoskeletal crosstalk.

    Current opinion in cell biology 2003, 15(1):67-72. PubMed Abstract | Publisher Full Text

    Return to text

57. Suzuki A, Ohno S: The PAR-aPKC system: lessons in polarity.

    Journal of cell science 2006, 119(Pt 6):979-987. PubMed Abstract | Publisher Full Text

    Return to text

58. Wodarz A: Establishing cell polarity in development.

    Nature cell biology 2002, 4(2):E39-44. PubMed Abstract | Publisher Full Text

    Return to text

59. Cau J, Hall A: Cdc42 controls the polarity of the actin and microtubule cytoskeletons through two distinct signal transduction pathways.

    Journal of cell science 2005, 118(Pt 12):2579-2587. PubMed Abstract | Publisher Full Text

    Return to text

60. Etienne-Manneville S, Hall A: Cdc42 regulates GSK-3beta and adenomatous polyposis coli to control cell polarity.

    Nature 2003, 421(6924):753-756. PubMed Abstract | Publisher Full Text

    Return to text

61. Harris TJ, Peifer M: aPKC controls microtubule organization to balance adherens junction symmetry and planar polarity during development.

    Developmental cell 2007, 12(5):727-738. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

62. Grabham PW, Seale GE, Bennecib M, Goldberg DJ, Vallee RB: Cytoplasmic dynein and LIS1 are required for microtubule advance during growth cone remodeling and fast axonal outgrowth.

    J Neurosci 2007, 27(21):5823-5834. PubMed Abstract | Publisher Full Text

    Return to text

63. Zheng Y, Wildonger J, Ye B, Zhang Y, Kita A, Younger SH, Zimmerman S, Jan LY, Jan YN: Dynein is required for polarized dendritic transport and uniform microtubule orientation in axons.

    Nature cell biology 2008, 10(10):1172-1180. PubMed Abstract | Publisher Full Text | PubMed Central Full Text

    Return to text

64. Allen MJ, Drummond JA, Moffat KG: Development of the giant fiber neuron of Drosophila melanogaster.

    J Comp Neurol 1998, 397(4):519-531. PubMed Abstract | Publisher Full Text

    Return to text

65. Freeman M: Reiterative use of the EGF receptor triggers differentiation of all cell types in the Drosophila eye.

    Cell 1996, 87(4):651-660. PubMed Abstract | Publisher Full Text

    Return to text

66. Tanouye MA, Wyman RJ: Motor outputs of giant nerve fiber in Drosophila.

    J Neurophysiol 1980, 44(2):405-421. PubMed Abstract | Publisher Full Text

    Return to text

67. Gorczyca M, Hall JC: Identification of a cholinergic synapse in the giant fiber pathway of Drosophila using conditional mutations of acetylcholine synthesis.

    J Neurogenet 1984, 1(4):289-313. PubMed Abstract | Publisher Full Text

    Return to text

68. Parks AL, Cook KR, Belvin M, Dompe NA, Fawcett R, Huppert K, Tan LR, Winter CG, Bogart KP, Deal JE, et al.: Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster genome.

    Nat Genet 2004, 36(3):288-292. PubMed Abstract | Publisher Full Text

        

On a cold day in January, Dan Stoicescu, a millionaire living in Switzerland, became the second person in the world to buy the full sequence of his own genetic code.

He is also among a relatively small group of individuals who could afford the $350,000 price tag.

Mr. Stoicescu is the first customer of Knome, a Cambridge-based company that has promised to parse his genetic blueprint by spring. A Chinese executive has signed on for the same service with Knome’s partner, the Beijing Genomics Institute, the company said.

Scientists have so far unraveled only a handful of complete human genomes, all financed by governments, foundations and corporations in the name of medical research. But as the cost of genome sequencing goes from stratospheric to merely very expensive, it is piquing the interest of a new clientele.

“I’d rather spend my money on my genome than a Bentley or an airplane,” said Mr. Stoicescu, 56, a biotechnology entrepreneur who retired two years ago after selling his company. He says he will check discoveries about genetic disease risk against his genome sequence daily, “like a stock portfolio.”

But while money may buy a full readout of the six billion chemical units in an individual’s genome, biologists say the superrich will have to wait like everyone else to learn how the small variations in their sequence influence appearance, behavior, abilities, disease susceptibility and other traits.

“I was in someone’s Bentley once — nice car,” said James D. Watson, the co-discoverer of the structure of DNA, whose genome was sequenced last year by a company that donated the $1.5 million in costs to demonstrate its technology. “Would I rather have my genome sequenced or have a Bentley? Uh, toss up.”

He would probably pick the genome, Dr. Watson said, because it could reveal a disease-risk gene that one had passed on to one’s children, though in his case, it did not. What is needed, he said, is a “Chevrolet genome” that is affordable for everyone.

Biologists have mixed feelings about the emergence of the genome as a luxury item. Some worry that what they have dubbed “genomic elitism” could sour the public on genetic research that has long promised better, individualized health care for all. But others see the boutique genome as something like a $20 million tourist voyage to space — a necessary rite of passage for technology that may soon be within the grasp of the rest of us.

“We certainly don’t want a world where there’s a great imbalance of access to comprehensive genetic tests,” said Richard A. Gibbs, director of the human genome sequencing center at Baylor College of Medicine. “But to the extent that this can be seen as an idiosyncratic exercise of curious individuals who can afford it, it could be quite a positive phenomenon.”

It was the stream of offers from wealthy individuals to pay the Harvard laboratory of George M. Church for their personal genome sequences that led Dr. Church to co-found Knome last year (most people pronounce it “nome,” though he prefers “know-me”).

“It was distracting for an academic lab,” Dr. Church said. “But it made me think it could be a business.”

Scientists say they need tens of thousands of genome sequences to be made publicly available to begin to make sense of human variation.

Knome, however, expects many of its customers to insist on keeping their dearly bought genomes private, and provides a decentralized data storage system for that purpose.

Mr. Stoicescu said he worried about being seen as self-indulgent (though he donates much more each year to philanthropic causes), egotistical (for obvious reasons) or stupid (the cost of the technology, he knows, is dropping so fast that he would have certainly paid much less by waiting a few months).

But he agreed to be identified to help persuade others to participate. With only four complete human genome sequences announced by scientists around the world — along with the Human Genome Project, which finished assembling a genome drawn from several individuals at a cost of about $300 million in 2003 — each new one stands to add considerably to the collective knowledge.

“I view it as a kind of sponsorship,” he said. “In a way you can also be part of this adventure, which I believe is going to change a lot of things.”

Mr. Stoicescu, who has a Ph.D. in medicinal chemistry, was born in Romania and lived in the United States in the early 1990s before founding Sindan, an oncology products company that he ran for 15 years. Now living with his wife and 12-year-old son in a village outside Geneva, he describes himself as a “transhumanist” who believes that life can be extended through nanotechnology and artificial intelligence, as well as diet and lifestyle adaptations. His genome sequence, he reasons, might give him a better indication of just what those should be. Last fall, Mr. Stoicescu paid $1,000 to get a glimpse of his genetic code from deCODE Genetics. That service, and a similar one offered by 23andMe, looks at close to a million nucleotides on the human genome where DNA is known to differ among people.

But Mr. Stoicescu was intrigued by the idea of a more complete picture. “It is only a part of the truth,” he said. “Having the full sequence decoded you can be closer to reality.”

How close is a matter of much debate. Knome is using a technology that reads the genome in short fragments that can be tricky to assemble. All of the existing sequencing methods have a margin of error, and the fledgling industry has no agreed-on quality standards.

Knome is not the only firm in the private genome business. Illumina, a sequencing firm in San Diego, plans to sell whole genome sequencing to the “rich and famous market” this year, said its chief executive, Jay Flatley. If competition drives prices down, the personal genome may quickly lose its exclusivity. The nonprofit X Prize Foundation is offering $10 million to the first group to sequence 100 human genomes in 10 days, for $10,000 or less per genome. The federal government is supporting technology development with an eye to a $1,000 genome in the next decade.

But for now, Knome’s prospective customers are decidedly high-end. The company has been approached by hedge fund managers, Hollywood executives and an individual from the Middle East who could be contacted only through a third party, said Jorge Conde, Knome’s chief executive.

“I feel like everyone’s going to have to get it done at some point, so why not be one of the first?” said Eugene Katchalov, 27, a money manager in Manhattan who has met with Mr. Conde twice.

Mr. Stoicescu, who wants to create an open database of genomic information seeded with his own sequence, hopes others will soon join him.

A few days after he wired his $175,000 deposit to the company, a Knome associate flew in from Cambridge to meet him at a local clinic.

“What the heck am I doing?” Mr. Stoicescu recalls wondering. “And how many children in Africa might have been fed?”

Then he offered up his arm and gave her three test tubes of his blood.

        

BOSTON — Is Esther Dyson, the technology venture capitalist who is training to be an astronaut, genetically predisposed to a major heart attack?

Does Steven Pinker, the prominent psychologist and author, have a gene variant that raises his risk of Alzheimer’s, which his grandmother suffered from, to greater than 50 percent?

Did Misha Angrist, an assistant professor at Duke University, inherit a high risk of breast cancer, which he may have passed on to his young daughters?

On Monday, they may learn the answers to these and other questions — and, if all goes according to plan, so will everyone else who cares to visit a public Web site, www.personalgenomes.org. The three are among the first 10 volunteers in the Personal Genome Project, a study at Harvard University Medical School aimed at challenging the conventional wisdom that the secrets of our genes are best kept to ourselves.

The goal of the project, which hopes to expand to 100,000 participants, is to speed medical research by dispensing with the elaborate precautions traditionally taken to protect the privacy of human subjects. The more genetic information can be made open and publicly available, nearly everyone agrees, the faster research will progress.

In exchange for the decoding of their DNA, participants agree to make it available to all — along with photographs, their disease histories, allergies, medications, ethnic backgrounds and a trove of other traits, called phenotypes, from food preferences to television viewing habits.

Including phenotypes, which most other public genetic databases have avoided in deference to privacy concerns, should allow researchers to more easily discover how genes and traits are linked. Because the “PGP 10,” as they call themselves, agreed to forfeit their privacy, any researcher will have a chance to mine the data, rather than just a small group with clearance.

The project is as much a social experiment as a scientific one. “We don’t yet know the consequences of having one’s genome out in the open,” said George M. Church, a human geneticist at Harvard who is the project’s leader and one of its subjects. “But it’s worth exploring.”

A new federal law prohibits health insurers and employers from discriminating against individuals on the basis of their genetic profile. But any one of the PGP 10 could be denied life insurance, long-term care insurance or disability insurance, with no legal penalty. And no law can bar colleagues from raising an annoyed eyebrow at a PGP participant who, say, indulges in a brownie after disclosing on the Internet that she is genetically predisposed to diabetes.

Then there is the matter of potential recrimination — from siblings, parents and children who share half of the participants’ genes and did not necessarily agree to display them in public. Prospective participants are advised to consult with first-degree relatives, but except for identical twins, their consent is not required. Some volunteers are worried about their hurting their teenagers’ dating prospects.

“A potential boyfriend could look at my genome and say, ‘I don’t know if this relationship is meant to be,’ ” said John Halamka, a participant and the chief information officer of Harvard Medical School, who has a 15-year-old daughter. (His daughter, he said, told him that if a suitor did that, “I wouldn’t want them as a boyfriend anyway.”)

Because of the known and unknown risks, Dr. Church required the first 10 participants to demonstrate the equivalent of a master’s degree in genetics. Most are either investors or executives in the biomedical industry, or else teach or write about it, so they may have a financial interest in encouraging people to part with their genetic privacy.

The project has drawn criticism from scientists and bioethicists who caution that even its highly educated volunteers cannot understand the practical and psychological risks of disclosing information long regarded as quintessentially private.

“I’m concerned that this could make it seem easy and cool to put your information out there when there is still a lot of stigma associated with certain genetic traits,” said Kathy Hudson, director of the Genetics and Public Policy Center at Johns Hopkins University. “There will be new uses of this data that people can’t anticipate — and they can’t do anything to get it back.”

For now, the PGP, which is privately funded, is sequencing only the fraction of participants’ genomes thought to have the most influence over disease, behavior and physical traits. But the question of how much value to place on genetic privacy has taken on more urgency as the technology for sequencing an entire human genome accelerated and the price has plummeted to as low as $5,000, so that it may soon be possible for everyone to possess their own genetic readout.

Sequencing a human genome — the six billion letters of genetic code containing the complete inventory of the traits we inherited from our parents — cost over $1 million just two years ago.

The two scientists whose full genomes were sequenced in the name of research both made them public. But they differ on whether the practice should be widely recommended.

“I put mine out there, but I’m 80,” said James D. Watson, the chancellor emeritus of Cold Spring Harbor Laboratory and co-discoverer of the structure of DNA. “Randomly putting up young people’s genomes could cause individual harm, simply because there will be so many mistakes. We don’t know enough yet to interpret them.”

J. Craig Venter, a pioneer in human genome sequencing, said his nonprofit institute planned to sequence several dozen human genomes by the end of next year and to deposit the information in the public domain along with phenotype information in a model similar to that of the PGP. He said he had already heard from thousands of volunteers.

“If they want privacy we tell them to go somewhere else,” Dr. Venter said. “To truly understand humans we need a huge data set of 10,000 complete genomes, and the data needs to be open to everyone for interpretation.”

Besides, promises of privacy may be impossible to keep, given the extraordinary identifying properties of DNA. Over the last three years, more than a half-million people who participated in over 100 publicly financed genetic studies on traits like schizophrenia and drug addiction were promised that their anonymity would be protected. But last month, after a paper in a scientific journal described how an individual’s profile could be identified even when it was aggregated with hundreds of others, the National Institutes of Health abruptly restricted access to the data.

There are some signs that the reflex to protect genetic privacy may be shifting. On the Web site of 23 and Me, a company that markets a $400 minisnapshot of traits from risk of heart disease to ear wax type, some customers use pseudonyms to discuss their results, while others include links with their contact information.

And Sergey Brin, the co-founder of Google, recently revealed on his blog that he learned he has a considerably higher than average risk of developing Parkinson’s disease, which was diagnosed in his mother several years ago. (Mr. Brin is the husband of Anne Wojcicki, a co-founder of 23 and Me.)

“There are costs to keeping things secret,” Mr. Brin said in an interview. “There’s a much better chance that you will learn something useful if you are not trying to hide it.”

Still, it may depend on what “it” is.

As the PGP 10 gathered Sunday at Harvard Medical School in Boston to receive the first batch of their genetic data, many said they were motivated by a desire to demystify genetics, which is often wrongly viewed as determining a person’s fate.

As the hour approached when they would be asked to reveal their data to the world, Dr. Pinker said he was still considering whether he wanted to learn of his Alzheimer’s risk, or if he would ask the researchers to withhold the data from himself and the public. Everyone, Dr. Church said, is given a chance to change their mind about going public up until the last minute, “but we try very hard in our screening process to choose the people who understand that it is better to have it all out there.”

Only about 1,300 of the 20,000 human genes have been so far linked to a particular trait, PGP researchers said.

Thus, even if Dr. Pinker chooses to remove from public view the chunk of DNA currently associated with Alzheimer’s risk, he is not necessarily protecting himself from future associations scientists may make about genetic data that may now seem innocuous enough to put on the Web.

Dr. Halamka, a PGP volunteer who found out Sunday afternoon that he has a gene variant that has been associated with childhood blindness, said he had no qualms about putting that, and all of his other information, online. Since he is not blind, and neither is his 15-year-old daughter, the project’s researchers told him it seemed likely that something in his genetic makeup was compensating for the defect.

Still, he asked whether it was associated with multiple sclerosis, which his father has. “My daughter,” Dr. Halamka said, “will be asking questions.”

What happens to the PGP, Dr. Church said, may serve as a litmus test for the fears of sharing genetic data, in an era when everyone’s inborn imperfections are becoming more identifiable. If this group is tracked “like major league baseball players, everyone will want to be like them,” he said. “If it runs into social hassles and financial hassles, then no one will.”

The volunteers will be given more information as the data is analyzed, and they may be asked to answer questions that might help researchers. But the only requirement is that they notify the project if they suffer any adverse effects from their participation.

Dr. Church said that information, too, will be made public.

        

ScienceDaily (Sep. 2, 2009) — Johns Hopkins engineers have devised innovative computer software that can sift through hundreds of genetic mutations and highlight the DNA changes that are most likely to promote cancer. The goal is to provide critical help to researchers who are poring over numerous newly discovered gene mutations, many of which are harmless or have no connection to cancer. According to its inventors, the new software will enable these scientists to focus more of their attention on the mutations most likely to trigger tumors.

A description of the method and details of a test using it on brain cancer DNA were published in the August 15 issue of the journal Cancer Research.

The new process focuses on missense mutations, meaning protein sequences that each possess a single tiny variation from the normal pattern. A small percentage of these genetic errors can reduce the activity of proteins that usually suppress tumors or hyperactivate proteins that make it easier for tumors to grow, thereby allowing cancer to develop and spread. But finding these genetic offenders can be difficult.

“It’s very expensive and time-consuming to test a huge number of gene mutations, trying to find the few that have a solid link to cancer,” said Rachel Karchin, an assistant professor of biomedical engineering who supervised the development of the computational sorting approach. “Our new screening system should dramatically speed up efforts to identify genetic cancer risk factors and help find new targets for cancer-fighting medications.”

The new computational method is called CHASM, short for Cancer-specific High-throughput Annotation of Somatic Mutations.

Developing this system required a partnership of researchers from diverse disciplines. Karchin and doctoral student Hannah Carter drew on their skills as members of the university’s Institute for Computational Medicine, which uses powerful information management and computing technologies to address important health problems, and collaborated with leading Johns Hopkins cancer and biostatistics experts from the university’s School of Medicine, its Bloomberg School of Public Health and the Johns Hopkins Kimmel Cancer Center.

The team first narrowed the field of about 600 potential brain cancer culprits using a computational method that would sort these mutations into “drivers” and “passengers.” Driver mutations are those that initiate or promote the growth of tumors. Passenger mutations are those that are present when a tumor forms but appear to play no role in its formation or growth. In other words, the passenger mutations are only along for the ride.

To prepare for the sorting, the researchers used a machine-learning technique in which about 50 characteristics or properties associated with cancer-causing mutations were given numerical values and programmed into the system. Karchin and Carter then employed a math technique called a Random Forest classifier to help separate and rank the drivers and the passengers. In this step, 500 computational “decision trees” considered each mutation to decide whether it possessed the key characteristics associated with promoting cancer. Eventually, each “tree” cast a vote: Was the gene a driver or a passenger?

“It’s a little like the children’s game of ‘Guess Who,’ where you ask a series of yes or no questions to eliminate certain people until you narrow it down to a few remaining suspects,” said Carter, who earned her undergraduate and master’s degrees at the University of Louisville and served as lead author of the Cancer Research paper. “In this case, the decision trees asked questions to figure out which mutations were most likely to be implicated in cancer.”

The election results—such as how many driver votes a mutation received—were used to produce a ranking. The genetic errors that collected the most driver votes wound up at the top of the list. The ones with the most passenger votes were placed near the bottom. With a list like this in hand, the software developers said, cancer researchers can direct more of their time and energy to the mutations at the top of the rankings.

Karchin and Carter plan to post their system on the Web and will allow researchers worldwide to use it freely to prioritize their studies. Because different genetic characteristics are associated with different types of cancers, they said the method can easily be adapted to rank the mutations that may be linked to different forms of the disease, such as breast cancer or lung cancer.

In addition to Karchin and Carter, the Johns Hopkins co-authors of the Cancer Research paper were Sining Chen, Leyla Isik, Svitlana Tyekucheva, Victor E. Velculescu, Kenneth W. Kinzler and Bert Vogelstein.

Funding for the research was provided by the National Cancer Institute, the Susan G. Komen Foundation, the Virginia and D. K. Ludwig Fund for Cancer Research and the National Institutes of Health.

        

Whether they wanted to or not, many Americans learned a lot about the state of their health care this year.

High unemployment, the swine-flu pandemic and growing political momentum for a health-system overhaul made 2009 a high-stakes year for health care, with 2010 poised to carry on a few big themes.

What to do about spiraling costs dominated the discussion at kitchen tables and bargaining tables across the country this year. Here are the key developments over the last year and what may be in store in each area in 2010:

 
   Job Losses, Coverage Continuation 

Studies suggest that for every 1% rise in the unemployment rate, about 1.1 million more people lose their health insurance. During a year of ongoing recession and a national unemployment rate that topped 10%, more Americans were exposed to Cobra, the continuation of employer-sponsored coverage that workers in businesses with at least 20 employees can opt for after a job loss.

As part of the stimulus bill Congress passed early in the year, workers laid off between Sept. 1, 2008 and Dec. 31, 2009 were eligible for nine months of premium assistance that defrayed the cost of these expensive but comprehensive health plans.

Legislation that would extend these subsidies is pending. That could be good news for people struggling to hold onto their coverage when they don't have any income.

Earlier this year, a study found that Cobra charges eat up the bulk of monthly unemployment checks, putting Cobra benefits out of reach for many. The average monthly Cobra premium for family coverage, $1,069, consumed 84% of the average monthly unemployment check, which was $1,287, according to Families USA, a national advocacy group.

Even those who held onto their job-based coverage faced higher costs. Despite seeing a relatively moderate 5% average increase in family premiums this year, premiums have jumped a total of 131% compared with 38% growth in workers' wages and a 28% rise in general inflation since 1999, according to the Kaiser Family Foundation.

 
   Swine Flu Tests Vaccine Production 

It was the year of two flu vaccines. In April, a deadly flu outbreak in Mexico put U.S. public-health officials on alert as a new strain of the virus identified as H1N1 swept the globe. Authorities moved swiftly to create a pandemic flu vaccine aimed for mass distribution in the fall. But manufacturing problems led to delays and supply shortages, exposing the flaws of the lengthy egg-based vaccine-production method that's still in use and generating renewed interest in cell-based technologies.

Swine flu was pervasive among children at summer camps and spread in a second wave when kids went back to school in late August. Demand for the new H1N1 vaccine ran high. As the first vaccine doses trickled out in October, the Centers for Disease Control and Prevention limited them to high-risk groups such as children and pregnant women. Swine flu went on to take a disproportionate toll on the young and middle-aged while largely sparing the elderly.

Many Americans who wanted a vaccine grew frustrated when they couldn't find one or had to stand in long lines. Meanwhile, distribution of the annual seasonal flu vaccine got off to an earlier-than-usual start.

The good news: The swine-flu vaccine remains the best protection against the illness and its potential complications, and many states are now dropping restrictions so it's available to anyone who wants it, the CDC said. Supply is now plentiful in many areas, with about 100 million doses on hand.

Almost all the circulating flu strains are still the H1N1 variety, and flu season lasts until May, providing a window of opportunity to get vaccinated. Several national drugstore chains are offering swine-flu vaccines. Walgreen is offering H1N1 vaccines in 27 states, with plans to expand availability to all 50 states by the end of December.

Even though swine-flu activity appears to be waning, health officials caution against complacency. They're concerned that the U.S. could see a resurgence of cases over the winter, a pattern that was a hallmark of the 1957 flu pandemic.

Swine flu has sickened an estimated 22 million Americans, hospitalized about 98,000 and killed 4,000, including 232 children, since it was first identified in April.

 
   Health-Overhaul Debate Turns Up The Volume 

President Barack Obama began shifting his health-reform agenda, his top domestic priority, into high gear in the spring, ushering in a summer of heated debate about what a remake of the health-care system should aim to accomplish and how it should be paid for.

Obama set broad parameters but let Congress tackle the details. He wants to expand coverage to millions of uninsured Americans, contain cost growth and improve health-care quality without adding to the budget deficit. Obama repeatedly points out that the U.S. is the only wealthy nation in the world that doesn't have universal coverage, yet it spends nearly twice as much as its international peers on health care.

With rapid cost growth weighing on businesses and consumers alike, this year's reform talks gained momentum. Last month, AARP and the American Medical Association, which opposed the 1993 reform attempt and the creation of Medicare in the 1960s, endorsed the House of Representatives' health-overhaul bill.

The House narrowly passed its bill on Nov. 7 with a vote of 220 to 215. The Senate has been debating its version since the beginning of December and early Monday morning cleared a key procedural hurdle on a 60-40 vote that keeps Democrats on track for a final vote before Christmas. The two bills then would have to be reconciled before Obama could sign one into law.

In the Senate, a proposal to include a public plan option to compete with private insurers has been dropped in recent negotiations, and the idea of opening Medicare to people age 55 to 64 is gone as well. Among the remaining disagreements that threaten passage, one of the biggest concerns is whether any federal funds could be used for abortion coverage.

 
   More Skepticism About Diagnostics, Treatments 

In 2009, consumers began comparing drugs, treatments, hospitals and doctors when possible as they looked for the least risk and greatest potential benefits for their overall health and the best value for their money, said John Santa, director of the Consumer Reports Health Ratings Center in Yonkers, N.Y.

"Historically, we'll look back on this and say this was the first time Americans began to have to deal with the reality of comparing health services, and that's big," he said.

Despite the clamor for evidence-based medicine, which promotes health care that is shown to be most clinically effective in peer-reviewed studies, many Americans were outraged when an independent research group suggested that annual mammograms for women in their 40s often did more harm than good.

The U.S. Preventive Services Task Force touched off a firestorm when it called for a more individualized approach to routine screening for 40-something women at average risk of breast cancer, raising questions about how to better communicate with the public about complex health information.

That same week, the American College of Obstetricians and Gynecologists relaxed its guidelines for routine cervical-cancer screening among young women but caused nowhere near the same furor.

"If people start cutting back on those things that are not so effective, we're talking about keeping dollars in the system that can be used on things that are effective," said Marge Ginsburg, executive director of the Center for Healthcare Decisions, a nonprofit in Sacramento, Calif.

This year also continued a trend of trying to root out conflicts of interest between doctors and drug and medical-device companies, but there's a long way to go, said Patrick Malone, a Washington-based attorney, patient-safety advocate and author of "The Life You Save."

"You've got to ask a lot of questions about your doctor to make sure you're in the right hands," he said. "Nobody in medicine wants to admit that A, they have a conflict, or B, it would influence them in any way."

"Part of this health-care reform debate has helped to educate people about the problems of overtreatment, and that's a good thing for people to start to be a little skeptical," Malone added.

 
   Doctors In The Spotlight 

Unlike the last health-reform debate 15 years ago, doctors were no longer exempt from scrutiny this year, said Dr. Howard Brody, director of the Institute for the Medical Humanities at the University of Texas Medical Branch at Galveston.

"We started to get some information coming to the forefront...that highlighted the role of the physician in driving up health-care costs," he said. "2010 could end up being the year when we grapple with this problem."

"Medicine has long been able to say 'Don't let anyone come between you and your doctor.' The doctor is on a pedestal. Whatever advice the doctor gives you must be right," Brody said. "I think we're starting to see that's just too simple."

An influential piece written by Dr. Atul Gawande in the June 1 edition of the New Yorker magazine highlighted big disparities in medical spending between two Texas towns, one of which cultivated a business culture around the practice of medicine.

"The doctors who practice in higher-cost regions, in higher-cost specialties sooner or later are going to have their feet held to the fire," Brody said. "If we ever in the future are going to have both a high-quality and a cost-effective health-care system, doctors are going to have to look themselves in the mirror and look at their own practices and their own prescribing patterns."

This year's debate also brought attention to a growing shortage of primary-care physicians and pilot programs that are reorienting the delivery system around patients' need for more coordinated and continuous care across settings.

        

You think you’re learning everything you can in your biology and anthropology classes, but textbook editors simply don’t have all the space they need to give you the full story of your body. Some of the facts below are trivial, some are ancient history, and some of them may very well save your life one day. So read up, and enjoy this wild and whacky anatomical analysis.

Unusual Facts

You’ll probably wonder why you never heard these cool facts in biology class before. From hangover cures to exploding head syndrome, these tidbits are must-knows.

1. Every person has a unique tongue print:…just like our fingers!
2. Eating fruits and vegetables may help the human body make its own aspirin: People who intake benzoic acid, a natural substance in fruits and vegetables, make their own salicylic acid, the key component that gives aspirin its anti-inflammatory and pain-relieving properties.
3. Baking soda can whiten teeth, garlic can help treat athlete’s foot, and honey can soothe a hangover.
4. Facebook may be good for your health: Studies show that staying in touch with family and friends can ward off memory loss and help you live longer.
5. Exploding head syndrome is real (but rare): The American Sleep Association explains that a person with exploding head syndrome experiences a loud, indecipherable noise that seems to originate from inside the head.
6. Regular exercise can lower a woman’s cancer risk, but only if she’s getting enough sleep: Check out the link to see what the National Cancer Institute has to say about this important fact.
7. Body position affects memory: Memories are highly embodied in our senses. A scent or sound may evoke a distant episode from one’s childhood. The connections can be obvious (a bicycle bell makes you remember your old paper route) or inscrutable.
8. Your bones can self-destruct: In addition to supporting the bag of organs and muscles that is our body, bones help regulate our calcium levels. If the element is in short supply, certain hormones will cause bones to break down, upping calcium levels in the body until the appropriate extracellular concentration is reached.
9. Your brain has a huge appetite: Though it makes up only 2 percent of our total body weight, the brain demands 20 percent of the body’s oxygen and calories.
10. Puberty reshapes the brain: Why is adolescence so emotionally unpleasant? Hormones like testosterone actually influence the development of neurons in the brain, and the changes made to brain structure have many behavioral consequences. Expect emotional awkwardness, apathy and poor decision-making skills as regions in the frontal cortex mature.

All About…Weight

Here are some great facts you should definitely know about how your body metabolizes and stores fat.

11. Weight really is genetic: But, a genetic predisposition isn’t necessarily a life sentence, experts say. Exercising regularly can offset the risk of obesity.
12. Some people just have more fat cells: While you can’t reduce your total number of fat cells, there are things you can do to keep them small. (See next point.)
13. You can change your metabolism: Gaining as little as 11 pounds can slow metabolism and send you spiraling into a vicious cycle: As you gain more fat, it becomes harder to lose it. But, scientists say physical activity can raise your metabolism back up to fat-blasting levels.
14. Stress fattens you up: The most direct route is the food-in-mouth syndrome: Stressful circumstances (your bank account, your boss) spark cravings for carbohydrate-rich snack foods, which in turn calm stress hormones.
15. Your mom’s pregnancy sealed your fate: Science says sugary and fatty foods, consumed even before you’re born, can wreak havoc on your future relationship  with fat.
16. Sleep more, lose more: University of Chicago researchers reported that sleep deprivation upsets our hormone balance, triggering both a decrease in leptin (which helps you feel full) and an increase of ghrelin (which triggers hunger).
17. Your spouse’s weight matters: Research shows that weight gain and loss can be, well, contagious. A study in the New England Journal of Medicine suggests that if one spouse is obese, the other is 37 percent more likely to become obese too. The researchers concluded that obesity seems to spread through social networks.
18. A virus can cause obesity: Adenoviruses are responsible for a host of ills, from upper respiratory tract problems to gastrointestinal troubles. It also seems to increase the number of fat cells in the body as well as the fat content of these cells.
19. Cookies really are addictive: When subjects at Monell Chemical Senses Center in Philadelphia were shown the names of foods they liked, the parts of the brain that got excited were the same parts activated in drug addicts.
20. Pick a diet, any diet: What really matters is your ability to moderate your intake of food. Feel free to use your favorite full-fat salad dressing, but your lettuce shouldn’t be swimming in it. You can cut carbs, fats, or just calories in general.

All About…Your Heart

It’s a sad fact that heart disease is the number one killer in the United States; what most people don’t realize is how preventable it is. Learn these facts, then do your part to protect your heart.

21. Laughter is therapeutic: Watching a funny movie for even 15 minutes can increase your blood flow. Remember to laugh every day—it can keep your heart happy and healthy.
22. Chest pain isn’t the only sign of a heart attack: Symptoms for most heart attacks include mild chest pain, some shoulder discomfort, or shortness of breath. Other signs can be nausea, lightheadedness, or breaking out in a cold sweat
23. If you’re over 20, you should know your cholesterol level: If it’s high, there are treatments (including medication and exercises) that can help. You should also get your blood pressure and your blood sugar levels checked regularly.
24. Obesity often leads to type 2 diabetes: Research has shown that eating more fruits, vegetables, and fiber can actually change the blood’s sensitivity to insulin within as little as two weeks.
25. Walking can save your life: A recent study found that a sedentary 40-year-old woman who begins walking briskly half an hour a day, four days a week, can enjoy almost the same low risk of heart attack as a woman who has exercised regularly her entire life.
26. Children can suffer from hypertension, too: About five out of every 100 children have higher than normal blood pressure.

You Have Super Powers!

Believe it or not, we’ve all got a little Clark Kent in us somewhere!

26. Super strength: Your fight-or-flight capacities point toward fighting, you can do the things you’ve heard of in the news, like lift cars off of your loved ones or push 600-pound boulders out of your way.
27. Supersonic hearing: Echolocation–it’s the way people with visual impairment continue to do amazing things. A prime example of this hidden sensory super power is Daniel Kish, a mountain biker who has been completely blind his whole life. He bikes better and faster than most people with vision, all by using sound to mentally paint a picture of the world around him. He does it so fast he can avoid trees, boulders and bears while speeding down the side of a mountain.
28. Super memory: Your brain technically has the ability to store every single thing you’ve ever seen or heard or experienced.
29. Super pain threshold: In moments of shock and trauma, your brain flips off pain like a switch. Ask somebody like Amy Racina, who fell off a cliff, landing six stories below, shattering her knee and breaking her hip. Not feeling more than minor pain, even with broken bone jutting out from her skin, she dragged herself until she found help. It was only at the point when she was being loaded safely into a helicopter that the pain returned.
30. Time manipulation: How fast time moves for you is literally all in your head. Experts say it’s because your brain has two modes of experiencing the world, rational and experiential. Read more about these modes and how they could save your life.

All About…Skin

These fantastic (but not-so-appetizing) facts will have you thinking twice every time you look in the mirror.

31. Weight game: An average adult’s skin spans 21 square feet, weighs nine pounds, and contains more than 11 miles of blood vessels.
32. Sweating like a…human: The skin releases as much as three gallons of sweat a day in hot weather. The areas that don’t sweat are the nail bed, the margins of the lips, the tip of the penis, and the eardrums.
33. Take a bath: Body odor comes from a second kind of sweat— a fatty secretion produced by the apocrine sweat glands. The odor is caused by bacteria on the skin eating and digesting those fatty compounds.
34. The perfect crime: Some people never develop fingerprints at all. Two rare genetic defects, known as Naegeli syndrome and dermatopathia pigmentosa reticularis, can leave carriers without any identifying ridges on their skin.
35. Hold your breath: Globally, dead skin accounts for about a billion tons of dust in the atmosphere. Your skin sheds 50,000 cells every minute.
36. It can see: In blind people, the brain’s visual cortex is rewired to respond to stimuli received through touch and hearing, so they literally "see" the world by touch and sound.
37. Master race, my melanin!: White skin appeared just 20,000 to 50,000 years ago, as dark-skinned humans migrated to colder climes and lost much of their melanin pigment.
38. You see very, very white people: Albinos are often cast as movie villains, as seen in The Da Vinci Code, Die Another Day, The Matrix Reloaded, and—inexplicably—the 2001 flick Josie and the Pussycats. Robert Lima of Penn State suggests that people associate pale-skinned albinos with vampires and other mythical creatures of the night.
39. Who needs keys?: More than 2,000 people have radio frequency identification chips, or RFID tags, inserted under their skin. The tags can provide access to medical information, log on to computers, or unlock car doors.
40. You’re reading what?!: The Cleveland Public Library, Harvard Law School, and Brown University all have books clad in skin stripped from executed criminals or from the poor.

All About…Hygiene

Throughout history, humans have evolved to become cleaner, healthier, and fresher-smelling beings, but we’re certainly not sterile. These somewhat-disturbing (but very interesting) hygenic truths might have you showering more often than ever before.

41. We’re really dirty: The human body is home to some 1,000 species of bacteria. There are more germs on your body than people in the United States.
42. Be careful what you wash with: Antibacterial soap is no more effective at preventing infection than regular soap, and triclosan (the active ingredient) can mess with your sex hormones.
43. Pee like an Egyptian: Ancient Egyptians and Aztecs rubbed urine on their skin to treat cuts and burns. Urea, a key chemical in urine, is known to kill fungi and bacteria.
44. You might want to skip the fountain drink: There are more bacteria in ice machines at fast-food restaurants than in toilet bowl water.
45. Floor food isn’t good: There’s no “five-second rule” when it comes to dropping food on the ground. Bacteria need no time at all to contaminate food.
46. TV can kill: TV remotes spread antibiotic-resistant Staphylococcus, which contributes to the 90,000 annual deaths from infection acquired in hospitals.
47. Make sure your doctor washes his hands: It is now believed President James Garfield died not from the bullet fired by Charles Guiteau but because the medical team treated the president with manure-stained hands, causing a severe infection that killed him three months later.
48. Mount Soapo: Soap gets its name from the mythological Mount Sapo. Fat and wood ash from animal sacrifices there washed into the Tiber River, creating a rudimentary cleaning agent that aided women doing their washing.
49. Sorry, Mom: Up to a quarter of all women giving birth in European and American hospitals in the 17th through 19th centuries died of puerperal fever, an infection spread by unhygienic nurses and doctors.
50. Don’t forget to floss: The first true toothbrush, consisting of Siberian pig hair bristles wired into carved cattle-bone handles, was invented in China in 1498. But tooth brushing didn’t become routine in the United States until it was enforced on soldiers during World War II.

All About…Sex and Gender

Click on these links, and you’ll get a whole new take on gender, sexuality, and the complexity of human relationships.

51. What were they waiting for?: Life emerged on earth about 3.8 billion years ago, but sex did not evolve until more than 2 billion years later.
52. Why does it take two?: Scientists are not sure, since asexual reproduction is a better evolutionary strategy in some important ways.
53. Better for your liver than tylenol: Sex cures headaches. Endorphins released into our bloodstream when we have sex not only give us pleasure but also act as painkillers.
54. Coo-coo for cocoa: 70% of women would rather eat choclate than have sex.
55. Staying alive: Sex wards off heart attacks. Frequent sexual intercourse (twice or more per week) lowers your chance of a fatal heart attack.
56. It’s actually a cure-all: It also decreases pain from menstrual cramps and arthritis. It increases levels of endorphins and corticosteroids, raising pain thresholds.
57. Calm yourself: Sex reduces stress, so try to make time for it at least a few times a week.
58. What’s love got to do with it?: Sexual arousal and romantic love activate quite distinct areas of the brain—and love is clearly the more powerful. The latter turns on dopamine-rich regions linked with motivation, and falling in love is not unlike the rush of taking cocaine, hence the addictiveness of a new crush, and the withdrawal-like symptoms of love lost.
59. Curious?: Recent research suggests women may be “intrinsically bisexual,” and the higher their libido, the more they desire both sexes.
60. Asexuality 101: One percent of adults have zero interest in sex and have never felt sexually attracted to anyone at all.

All About…Mental Health

Mental health and sickness is a big part of being human. Though it’s been looked at and discussed with ridicule in previous generations, today’s world should know these important facts about human psychology and psychiatry.

61. Mental illness is more common than you think: The National Institute of Mental Health reports that One in four adults-approximately 57.7 million Americans-experience a mental health disorder in a given year.
62. Adults aren’t the only ones who suffer: The U.S. Surgeon General reports that 10 percent of children and adolescents in the United States suffer from serious emotional and mental disorders that cause significant functional impairment in their day-to-day lives at home, in school and with peers.
63. Mental illness strikes young: Mental illness usually strike individuals in the prime of their lives, often during adolescence and young adulthood. All ages are susceptible, but the young and the old are especially vulnerable.
64. It’s expensive: The economic cost of untreated mental illness is more than 100 billion dollars each year in the United States.
65. It’s treatable: Most people who live with serious mental illnesses can significantly reduce the impact of their illness and find a satisfying measure of achievement and independence.
66. Act fast: Early identification is of vital importance to treatment.
67. Don’t be bullied: Stigma erodes confidence that mental disorders are real, treatable health conditions.
68. Bipolar disorder is blind to gender: Over 10 million people in America have bipolar disorder, and the illness affects men and women equally.
69. Suicide is serious: Each year in the U.S., approximately 2 million U.S. adolescents attempt suicide, and almost 700,000 receive medical attention for their attempt.
70. Seasonal Affective Disorder (SAD) might be genetic: Many people with SAD report at least one close relative with a psychiatric condition, most frequently a severe depressive disorder (55 percent) or alcohol abuse (34 percent).

All About…Working Out

You already know how essential physical fitness is, and how important it is to take care of your body. But we’re willing to bet you didn’t know all of these bits of information about how to work out in a way that benefits you best.

71. Incline builds muscle: Get off the flats and into the hills. Walking uphill makes you stronger.
72. Treat your feet: Each step puts up to 1.5 times your body weight on your feet, which are shock absorbers that bear 60 tons of pressure every mile you travel.
73. Lub-dub: The more fit you are, the harder it is to reach your target heart rate.
74. Work it, girl: A 145-pound woman walking briskly (4 mph) burns 82 calories every 15 minutes.
75. Getting back in: If you’ve taken off more than one month from exercising, give yourself at least four weeks to regain endurance and strength. Start slow, and increase your time, distance and intensity gradually, even if you were previously well-trained.
76. Caffeine won’t kill: Despite popular thinking, caffeine doesn’t cause dehydration during exercise. In fact, the caffeine in a cup of tea or coffee improves speed and endurance.
77. Change it up: Running on a circular track puts stress on the outside of your foot. Alternate the direction you run every other day or every five laps.
78. Tums, anyone?: Strenuous exercise can cause acid reflux in healthy, conditioned athletes.
79. Go shopping: Keep the spring in your step. Replace your running shoes every 300 to 400 miles.
80. You’re not a camel: If you’re dehydrated even 5%, your metabolism can fall 20% to 30%. You’ll fire more easily and be irritable.

All about…Preventing Cancer

Cancer is the number two killer of people in the United States, but it doesn’t have to put you in the ground, too. Follow these healthy tips to treat your body to a longer, healthier life.

81. Serve sauerkraut at your next picnic: The fermentation process involved in making sauerkraut produces several other cancer-fighting compounds, including ITCs, indoles, and sulforaphane. To reduce the sodium content, rinse canned or jarred sauerkraut before eating.
82. Eat your fill of broccoli, but steam it rather than microwaving it: Broccoli is a cancer-preventing superfood, one you should eat frequently. But microwaving broccoli destroys 97 percent of the vegetable’s cancer-protective flavonoids.
83. Toast some Brazil nuts and sprinkle over your salad: They’re a rich form of selenium, a trace mineral that convinces cancer cells to commit suicide and helps cells repair their DNA.
84. Add garlic to everything you eat: Garlic contains sulfur compounds that may stimulate the immune system’s natural defenses against cancer, and may have the potential to reduce tumor growth. Studies suggest that garlic can reduce the incidence of stomach cancer by as much as a factor of 12!
85. Eat cantaloupe: Cantaloupe is a great source of carotenoids, plant chemicals shown to significantly reduce the risk of lung cancer.
86. Drink water: The amount of water women drink correlates to their risk of colon cancer, with heavy water drinkers reducing their risk up to 45 percent.
87. Get about 15 minutes of sunlight each day: Getting too little vitamin D may increase your risk of multiple cancers, including breast, colon, prostate, ovarian, and stomach, as well as osteoporosis, diabetes, multiple sclerosis, and high blood pressure.
88. Sprinkle scallions over your salad: A diet high in onions may reduce the risk of prostate cancer 50 percent. But the effects are strongest when they’re eaten raw or lightly cooked.
89. Make a batch of fresh lemonade or limeade: A daily dose of citrus fruits may cut the risk of mouth, throat, and stomach cancers by half, Australian researchers found.
90. Take advantage of your friends and family (they won’t mind): Men with high levels of stress and those with less satisfying contacts with friends and family members had higher levels of prostate-specific antigen (PSA) in their blood, a marker for the development of prostate cancer.

Weird Parts We Don’t Need

Evolution has gotten us pretty far, from our apelike ancestors to the upright walking, talking, complex thinking creatures we are today. But along the way, it left behind some traces of our previous">www.bloggingwv.com="">previous forms, like extrinsic ear muscles and tail bones.

91. Vomeronasal organ: A tiny pit on each side of the septum is lined with nonfunctioning chemoreceptors. They may be all that remains of a once extensive pheromone-detecting ability.
92. Wisdom teeth: Early humans had to chew a lot of plants to get enough calories to survive, making another row of molars helpful. Only about 5 percent of the population has a healthy set of these third molars.
93. Third eyelid: A common ancestor of birds and mammals may have had a membrane for protecting the eye and sweeping out debris. Humans retain only a tiny fold in the inner corner of the eye.
94. Male Nipples: Lactiferous ducts form well before testosterone causes sex differentiation in a fetus. Men have mammary tissue that can be stimulated to produce milk.
95. Body hair: Brows help keep sweat from the eyes, and male facial hair may play a role in sexual selection, but apparently most of the hair left on the human body serves no function.
96. Female vas deferens: What might become sperm ducts in males become the epoophoron in females, a cluster of useless dead-end tubules near the ovaries.
97. Fifth toe: Lesser apes use all their toes for grasping or clinging to branches. Humans need mainly the big toe for balance while walking upright.
98. Coccyx: These fused vertebrae are all that’s left of the tail that most mammals still use for balance and communication. Our hominid ancestors lost the need for a tail before they began walking upright.
99. Thirteenth Rib: Our closest cousins, chimpanzees and gorillas, have an extra set of ribs. Most of us have 12, but 8 percent of adults have the extras.
100. Extrinsic ear muscles: This trio of muscles most likely made it possible for prehominids to move their ears independently of their heads, as rabbits and dogs do. We still have them, which is why most people can learn to wiggle their ears.