[BioC] continous variable and limma

[grammat at csd.uoc.gr]

Hello All,

I have an dataset like the one below:

                         S1.CEL      S2.CEL      S3.CEL     S4.CEL      
   S5.CEL      S6.CEL       S7.CEL     S8.CEL       S9.CEL     S10.CEL  
    S11.CEL
1415670_at      3.147518    3.173544    3.169437    3.136441     
3.143318    3.149240    3.164243    3.058996    3.113238    3.189297    
  3.132998
1415671_at      3.416794    3.359792    3.355241    3.348604     
3.386615    3.370229    3.370467    3.377304    3.431597    3.378310    
  3.382426
1415672_at      3.387410    3.398707    3.417004    3.423049     
3.380673    3.397461    3.432254    3.421259    3.381972    3.395363    
  3.429707
1415673_at      3.122923    2.837839    3.024192    2.843976     
3.057093    2.912823    3.029993    2.878526    3.075537    2.897501    
  3.019125
1415674_a_at    3.233006    3.263457    3.257603    3.231753     
3.200804    3.177175    3.202201    3.231453    3.200878    3.222942    
  3.215434
1415675_at      3.097188    3.115366    3.162549    3.124800     
3.080499    3.057393    3.137030    3.136152    3.054494    3.083940    
  3.115158

and an index
           X genotype Orgaan tijdspunt
1 S1.CEL       WT  kidney    13
2 S2.CEL       WT  liver        13
3 S3.CEL       WT  spleen     13
4 S4.CEL       WT  lung        13
5 S5.CEL       WT  kidney     13
6 S6.CEL       WT  liver         13
...

Actually the design of my study is a little bit more complex:
I have more than 2 time points and i want to model time as a continous  
variable. I have 4 tissues and 2 mouse types.
Questions:
1. How could I find the the genes that significantly change over time  
separately in each tissue?
2. How could I statistically compare the slopes over time of the same  
gene between 2 or more tissue?