Figure 1.
Methods for embryonic kidney culture.
(A) The low-volume culture method described in this paper, drawn from the side and photographed from above. (B) Conventional, high volume culture either on a membrane on a Trowell grid (above) or on the membrane at the bottom of a well insert (below). The blue in (A) depicts PBS, the red in all diagrams depicts culture medium, yellow in (A) symbolizes the silicone ring and dark grey the embryonic kidney rudiments. The numbered divisions on the ruler are centimetres.
Figure 2.
Development of mouse kidney rudiments in conventional culture and on glass inside silicone rings.
(A–C) show phase contrast views of kidney rudiments grown for 4 days in silicone rings on glass in 85 µl, 120 µl and 200 µl respectively, while (D) shows a kidney grown in the conventional system (on a filter on a Trowell grid: the ‘noise’ in the background is the filter). On glass, the lowest volume, 85 µl, shows the best development, resulting in a larger kidney than the conventional system. (A′–D′) show kidneys grown in the same conditions as (A–D) but stained for basement membrane marker laminin (red) and the ureteric bud marker calbindin-D28k (green). (E) shows a quantitative analysis of area, nephron and bud tip numbers for each of these culture conditions. Error bars depict standard errors of the mean and are derived from at least 49 kidneys in total, from six different runs of the experiment, each run using between 6 and 18 kidneys. Scale bars = 500 µm.
Figure 3.
Normal segmentation of nephrons grown on filters (A, C, E, G) and in low volumes on glass (B, D, F, H).
(A, B) show kidneys stained for calbindin, which stains the whole ureteric bud; (C, D) show kidneys stained for Dolichos biflorus agglutinin (DBA), which stains only the bud stalk–the ‘missing’ tips visible in A,B but not in C,D are marked with yellow stars; (E, F) show kidneys stained for the WT1, which is expressed strongly in crescents that consist of developing podocytes: in these images, staining for Pax2, expressed in bud, condensates and early nephrons, is used to reveal the general structure of the rudiment and to place the developing podocytes in their anatomical context: higher power views are shown in E′ and F′, revealing the nuclear location of WT1. (G, H) show higher power views of kidneys stained for the ureteric bud and distal tubule marker, E-cadherin, and the proximal tubule marker megalin; the expression of each of these markers is similar in both culture systems. Scale bars = 50 µm; the scale bar shown in F applies to A–F, that shown in F′ to E′ and F′, and that shown in H applies to G and H.
Figure 4.
There is no significant difference after two days in rates of proliferation or cell death in nephrons formed in the two culture systems.
Proliferation was measured using BrdU incorporation and apoptosis by terminal end labelling (TUNEL); source images typical of those analysed quantitatively are shown for BrdU (A glass; C filter) and TUNEL (B glass; D filter). Over many such images, the numbers of BrdU or TUNEL-positive nuclei per unit area of visible nephron cross-section were counted. The data, from ten filter-grown kidneys and thirty one glass-grown ones, (E) show no evidence for difference between culture systems (a two-tailed Student's t-test yields p = 0.93 and p = 0.94 for BrdU and TUNEL respectively). Scale bar is 100 µm.
Figure 5.
Kidneys grown in the low volume glass system show normal responses to known morphogens.
(A, B) show the enhanced development of kidneys grown in 100 ng/ml GDNF in the low volume glass system and on conventional filters, stained for laminin (red) and calbindin-D29k (green). (C, D) show kidneys grown in control medium and (E, F) show the decreased development of kidneys cultured in the presence of 1 nM TGFβ. (G) shows the effects of these molecules on ureteric bud (UB) branching, in the low volume and the filter systems and (H) shows their effects on nephron number. Each bar represents data from between 10 and 71 kidneys. All values in (G) and (H) differ significantly from their respective controls, by a two-tailed Student's t-test assuming unequal variances, the weakest difference (that between nephrons in control and TGFβ, on glass) still having a p-value of 0.009. Scale bar 500 µm.
Figure 6.
Kidneys grown on glass develop organotypic features including cortico-medially zonation.
(A) Shows a time course of development to 240 h (10 d). From about 108 h (4.5 d), the kidney spreads out enough that it begins to divide into two zones, an outer medullary zone that features many nephrons and an inner medulla that contains more elongated tubules (mainly collecting ducts at this stage). By 240 h (10 d), this effect has become more marked. The last panel of (A) shows a kidney rudiment cultured on a filter for 240 h: the organ occupies less area and there is less evidence of corticomedullary zonation. The ‘grain’ in the photograph is an optical effect of the filter pores, which are beneath the kidney; their absence in bright field imaging is another advantage of the glass system. (B) A high-power view of part of a 240 h kidney shows that nephrons (red) are arranged organotypically, with the glomerulus (‘Glom’) and proximal convoluted tubule (‘PCT’) in the cortex and a loop of Henle (‘LoH’) extending down into the medulla, forming a tight hair-slide shape parallel to the collecting ducts (green). Green = calbindin, red = laminin.
Figure 7.
Kidneys grown in the low-volume glass system show less evidence of stress.
In (A) the samples labelled ‘short’ are of medium conditioned from day 0–2 and those labelled ‘long’ are from days 2–4. Spots marked ‘C’ are positive controls (to verify the detection kit). The other spots, of TIMP1, KC, JE and IFNγ represent these markers of cellular stress (see main text). (B) shows, quantitatively, the results of these molecules on the arrays, as means of three runs. The Y axis is % relative to the signal provided by the manufacturers' positive control. The right-most part of the graph, showing TIMP (also shown as a spot in A), is plotted to a different y axis. Although the data for each of the molecules, each individual array spot being derived from 3 pooled kidneys and 3 independent arrays being run for each condition, suggests a decrease on glass, the variation in measurement is so large that no one individual molecule has a change that can be regarded as significant at the conventional p≤0.05 (most have p≈0.1). As all molecules are markers for the same physiological state (cellular stress), however, their measurements can be pooled for a one-tailed, paired Student's t-test that uses all 15 experiment-control measurement pairs together (the 15 pairs coming from 5 target molecules x 3 runs each). Viewed this way, the data show that glass culture does produce a significant reduction in production of these stress proteins (p = 0.05). Elements on the spot arrays not shown in this graph showed no significant differences.
Figure 8.
Reduction of surface tension influences development of cultured kidney rudiments.
In (A) the reduction of surface tension by the addition of the surfactant DOPC to the medium is demonstrated; data represent means of a minimum of 18 measurements. (B and C) show the influence of the surfactant on ureteric bud branching and amount of nephrons, respectively, if compared to the control; each bar represents data from a minimum of 17 kidneys. The asterisk indicates significant difference (p = 3.4×10−6 for branches and 3.3×10−4 for nephrons) by a two-tailed Student's t-test assuming unequal variances.