Here we present the first genome-wide MCB-based study on 3D-preserved interphase nuclei derived from sperm (Fig. 1). Previously, comparable FISH-studies on sperm were performed on flattened nuclei with the known disadvantages of possible artifacts due to transformation of a spherical into a pancake-like object 11242526 or even on decondensed nuclei with DNA looping out 2728. As for probes, the ones used were: whole, or arm-specific chromosome paintings, or centromeric probes 23242526272829. We are aware of only one previous 3D-study on human sperm done by confocal microscope 22.

As summarized in Figures 2, 3, 4, 5, 6, 7, 8, 9 and Table 1 the statistical analysis revealed correlations between the investigated parameters central/peripheral, head/middle/tail, axial/non-axial, linear/non-linear and orientation of the chromosomal arms towards the sperm head, when analyzing the chromosomes by groups (see below).

The ISCN provides the subdivision of human chromosomes in sub-groups A through G 30. This is based on the chromosomal shape, size and centromeric position. In Figs. 3, 5, 7 and 9 clusters were formed from the obtained data and correlated with these chromosomal sub-groups.

A- and B-group chromosomes are primarily located in the periphery (Fig. 3). While A-group is equally orientated towards the tail and head of the sperm, B-group is more frequently found in the middle and head part of the sperm (Fig. 5 and data not shown). These results are in concordance with previous studies for #1 2529, #2 25 and #5 23. According to 29 the chromosome 2 is located more towards the tail of the sperm, which we could confirm by MCB.

Chromosomes of the C-group are positioned basically in the middle of sperm (Fig. 5). As visible in Fig. 3 C-group is localized preferentially in the periphery. Exceptions are here according to Fig. 3 chromosomes #6 and #12. It has to be noticed that for chromosome 6 already different data was reported 29, i.e. that chromosome 6 was observed to behave like all other chromosomes of the C-group as detected here. This discrepancy might be due to the fact that Zalenkaya and Zalensky (2004) 29 did their study on swollen and flattened slides and in parts with centromeric probes. On the other hand they report similar data as we found for chromosome 7 29. So further studies are necessary for a better understanding of the nuclear position of chromosome 6.

The D-group chromosomes can all be found in the central part of the sperm nucleus (Fig. 3). Also all three of them belong to the cluster which whereabouts are in the middle of the sperm (Fig. 5). Nonetheless, it is obvious that a tendency towards head or tail localization increases starting from #15, over #14 to #13. Hazzouri and coworkers found comparable data for chromosome 13, previously 22.

The E-group chromosomes #16 and #17 are localized in between middle and head and more towards the center of the sperm, as previously found by 29. Distribution of chromosome #18 essentially differs from other chromosomes of this group as this chromosome is of preference to find on periphery in tail or head of the sperm cell (Figs. 3 and 5), as previously reported by others 26.

The F-group chromosomes #19 and #20 can both mainly be found towards the sperm head (Fig. 5 and data not shown). However, for the distribution towards periphery and center chromosomes #19 (central position) and #20 (peripheral position) differ clearly from each other. A similar observation was the outcome of this study for the G-group chromosomes 21 (peripheral position) and 22 (central position). On the other hand, both G-group chromosomes are located in the middle of the sperm (Figs. 3 and 5).

Both gonosomes show a similar distribution, i.e. they are found in the periphery of the sperm. The X-chromosome is localized towards the middle and tail of the sperm (Fig. 5 and data not shown), while the Y-chromosome can be found in the middle part.

All data concerning chromosomal orientation and configuration in sperm are summarized in Figs. 7 and 9 and Tab. 1.

Of the A-group chromosomes #1 and #3 tend to be localized more axial than #2. Surprisingly, the longest human chromosome #1 tends to be configured more linear rather than #2 and #3. Also the B-group chromosomes have a more linear configuration. #5 tends to be axial orientated, #4 non-axial.

The C-group is arranged in a non-axial way, in general. Chromosomes #7 and #8 are exceptions here. Chromosomes #6 to #9 tend to be more linear than the others of the C-group. D-group chromosomes turned out to be located non-axial. Chromosomes #13 and #15 are arranged non-linear, #14 more linear.

The E-group falls again in two different clusters: chromosomes #16 and #17 are configured non-axial and non-linear, while chromosome #18 behaves the other way round. The same holds true for the F-group: #19 is non-axial and non-linear, #20 axial and linear arranged. In G-group similar things is to substitute by the same, even though the difference in axial/non-axial is not that expressed. Chromosome #22's configuration is more linear that of #21.

The gonosomes again behave very similar in terms of a more linear and neither expressed axial or non-axial behaviour.

A prevalence in orientation of p- or q-arm towards the sperm head could not be observed for half of the chromosomes (Tab. 1). Possible tendencies were observed for chromosomes #4, #6, #8, #13, #16, #20, #22 (p-arm) and #3, #5, #11, #12 and #15 (q-arm). However, as these results were obtained on 3 to 22 nuclei with axial orientated cells, they have to be considered as preliminary.

As summarized in Fig. 10 a direct correlation of position of the chromosomes (from center to periphery) with their size can be found for most of the chromosomes. Apart from 8 chromosomes (#1, #2, #6. #14, #18, #20, #21 and Y) all other closely adjoin to the straight line of regression according with their size. The correlation level is high (correlation coefficient = 0.952).

When arranging the smaller sized chromosomes (groups E, F, G and Y chromosome) then the gene density seems to have a more significant influence on the positioning of chromosomes in sperm. No such observation was possible for the larger chromosomes (results not shown). In Fig. 11 the smaller sized chromosomes are closely adjoined to the straight line of regression but comparing with the size dependence distribution relation has the opposite value. Chromosomes with a higher gene density are located in the center. Therefore, the Correlation Coefficient has negative value and equal to -0.983. The correlation level is high, however, chromosomes #20 and #22 do not completely fit.

Human sperm sample was collected in a sterile container after 3 days of sexual abstinence from a fertile, 30 year-old man with normal seminal parameters and normal karyotype. After liquefaction at room temperature, the samples was washed three times in 1 × phosphate-buffered saline by centrifugation (5 min at 2000 rpm) and fixed in fresh fixative (1:3 glacial acetic acid: methanol) 33.

S-FISH on interphase sperm cells was done as previously reported 11. 30 cells were evaluated per chromosome i.e. overall 720 interphase nuclei were analyzed. The same number of evaluated cells per chromosome was chosen in comparable previous studies 1129.

Analysis of chromosomal position in sperm included several parameters. The localization of chromosomes in periphery or center was determined. For that the sperm was divided into two spheres and 50% of the radius of cell was defined as central, the remainder as peripheral (Figs. 1, 2, 3). As the sperm axis can be determined in DAPI-staining (see 11 and Fig. 1a) it is also possible to define chromosomal localization as head, middle and tail parts of the sperm (Figs. 1; 4, 5). Moreover, it was determined if the chromosomes located along the longitudinal axis of the sperm or not. Chromosomes were defined as axial when the deviation was less the 45° and/or the chromosome was not entangled (Figs. 1; 6, 7). The latter was also registered by distinguishing linear and non-linear configuration of chromosomes (Figs. 1; 8, 9). Finally, the orientation concerning the short p- and long q-arms of sperm chromosomes towards the head of sperm was recorded (Fig. 1; Tab. 1).

Statistical analysis was performed using Student's t-test and One Way ANOVA (Analysis of Variance) to determine significant differences of chromosome's arrangement in sperm. Statistical significance was defined as p < 0.05.

Estimation of similarity in a position and orientations between various chromosomes was done with the application of cluster analysis (Figs. 3, 5, 7 and 9). The purpose of this algorithm is to correlate chromosomes within clusters, which are depicted as hierarchical trees. Linkage distances between chromosomes are computed based of Euclidean distances. This is simply the geometric distance in the multidimensional space. For analysis of dependence of chromosomes distribution on periphery and the center of sperm from the length and relative gene density of chromosomes, the regression analysis was used (Figs. 10, 11).