Hi, its probably a mechanical process that does it. And some papers on this by Nedelec et al. have shown in simulation and in vitro assembly that it might be the case. But the mechanics are complex involving mixtures of motors and microtubules. You can look up the following references for more:

Tanenbaum ME & Medema RH (2010) Review Mechanisms of Centrosome Separation and Bipolar Spindle Assembly. Dev. Cell 19: 797–806

Dinarina A, Pugieux C, Corral MM, Loose M, Spatz J, Karsenti E & Nédélec F (2009) Chromatin shapes the mitotic spindle. Cell 138: 502–13 Available at: http://www.ncbi.nlm.nih.gov/pubmed/19665972

Centrioles from at right angles during the S phase; the orientation most likely is due to how they were constructed in addition to serving functional purposes.

@mannen- i suppose the WHY = in order to generate 2 poles at the time of division and give rise to equal segregation of genomes (default in many somatic cells). HOW = a research question!

Actually there is very good evidence that the perpendicular configuration (engagement) regulates the duplication of centrioles (engaged centrioles cannot be duplicated; duplication once, and only once, per cell cycle is essential for proper bipolar mitosis). The disengagement of centrioles requires them to pass through mitosis (licensing step), so that in the subsequent S phase the disengaged centrioles can be duplicated. Each of the newly formed two mother/daughter pairs will then again be in the engaged configuration (to prevent further duplication) until they have gone though mitosis and each pair has been segregated to one of the two daughter cells.

You should check out some key papers on this:

Polo kinase and separase regulate the mitotic licensing of centriole duplication in human cells.

Tsou MF, Wang WJ, George KA, Uryu K, Stearns T, Jallepalli PV.

Dev Cell. 2009 Sep;17(3):344-54. doi: 10.1016/j.devcel.2009.07.015.

Centrosome duplication: of rules and licenses.

Nigg EA.

Trends Cell Biol. 2007 May;17(5):215-21. Epub 2007 Mar 26. Review.

Mechanism limiting centrosome duplication to once per cell cycle.

Tsou MF, Stearns T.

Nature. 2006 Aug 24;442(7105):947-51. Epub 2006 Jul 19.

Hello! I agree with Jens. I would like only to add that in some specialized cells that no longer divide, two centrioles can keep close to the perpendicular orientation.

Furthermore, duplication of centrosome begins in the late G1 phase before beginning of DNA replication. I prepared special study to confirm it . See in these publications :

Uzbekov R. E. (2007) Centriole duplication in PE (SPEV) cells starts before beginning of DNA replication. Biochemistry (Moscow) Supplement Series A: Membrane and Cell Biology, 2007, Vol. 1, No. 3, pp. 206–211.

Uzbekov R.E., and Alieva I.B. (2010) Centriole duplication or DNA replication – what starts earlier ? In: Cytoskeleton: Cell Movement, Cytokinesis and Organelles Organization (Editors S. Lansing and and T. Rousseau). Chapter 5. Nova Science Publishers, Inc.

I just want to draw your attention to one observation I published in the following paper (paper downloadable from my profile:)

Subhash C Juneja, Jan M van Deursen. 2005. A mouse model of familial oligoasthenoteratozoospermia. Human Reproduction (Impact Factor: 4.67). 05/2005; 20(4):881-93.

Please compare Figure 4A (in wild type mouse, centrioles are perpendicular to each other) and sperm arises from one centriole normally, whereas in Fig. 6B (mutant mouse), centrioles are not perpendicular and lies in a row, sperm tail is formed at an angle causing tail to make circle around the nucleus. within the sperm cell.

In this case gene mutation is the cause of altering angle of centrioles, such a possibility does exist. This observation will justify perpendicular nature of centrioles in sperm. at least.

Thanks