Biology 164 Laboratory Artificial Selection in Brassica, Part I

Biology 164 Laboratory

Artificial Selection in Brassica, Part I (Based on a laboratory exercise developed by Professor Bruce Fall, University of Minnesota)

I. Objectives

1. Gain familiarity with the process of artificial selection

2. Gain familiarity with some plant varieties in the genus Brassica that are the products of artificial selection

3. Conduct an artificial selection experiment involving a variety of Brassica rapa in which you will

a. quantify or assess the variability of one particular trait in a present generation of plants, and

b. attempt to change the genetic makeup of the next generation with respect to this trait so that, on average, the next generation exhibits the trait to a substantially greater degree than the present generation.

II. Introduction

A visit to a farm, supermarket, pet store or plant nursery will offer many examples of selective breeding of plants and animals by humans. Over hundreds and even thousands of years, humans have altered various species of plants and animals for our own use by selecting individuals for breeding that possessed certain desirable traits. This selective breeding process was continued for generation after generation. Often the products of such selective breeding are remarkable. Quite diverse domestic dog breeds, from chihuahuas and miniature poodles to Newfoundlands and Irish wolfhounds are all related to a common wild ancestor, the wolf (Canis lupus). Domestic chickens are all derived from the wild Jungle Fowl (Gallus gallus). As described in The Origin of Species, Darwin was particularly taken with the number of pigeon varieties such as tumblers, pouters, fantails and many others. All of these pigeon varieties were derived from wild Rock Doves (Columba livia) over the past 5,000 years.

One can find similar examples of selective breeding among plants, including those humans have bred for food. One plant group particularly important as a human food source is Brassica, a genus of plants in the mustard family (Brassicaceae). A number of nutritious and tasty vegetables have originated from three species of Brassica (Brassica rapa, B. oleracea and B. juncea). Some varieties have been bred for root production, others for leaves, flower buds or oil production. This group of plants is of great economic importance and hence a lot of research. The genetic relationships among the different forms have been thoroughly studied and are now fairly well understood. You may be surprised to learn that these familiar vegetables so different in appearance have the same species as a common ancestor. Centuries of artificial selection have produced greatly divergent forms. The following varieties originate from these wild species:

Artificial Selection in Brassica, Part I Page 1 Brassica oleracea – kale, cauliflower, broccoli, cabbage, Brussels sprouts, kohlrabi, collards, Savoy cabbage

Brassica juncea – leaf mustard, root mustard, head mustard and lots of other mustard varieties

Brassica rapa – turnip, Chinese cabbage, pak choi, rapid-cycling

Fig. 1: Four varieties of Brassica rapa.

The success of plant and animal breeders in dramatically changing the appearance of various lineages of organisms in a relatively short period of time is an obvious yet profound fact. This fact did not escape Charles Darwin as the first chapter of The Origin of Species concerns artificial selection by humans (“Variation under Domestication”). Darwin used many examples of selection by humans to help support the case for his proposed mechanism for the evolution of natural populations – natural selection.

To gain a better understanding of selection and inheritance, biologists have experimented with artificial selection involving a variety of traits in many different species of bacteria, yeast, plants and animals. The results, obtained in a relatively short period of time, are often impressive.

A general finding of these studies is that most variable traits in organisms respond to artificial selection. In other words, it is usually possible to increase or decrease the frequency or average value of a trait in a lineage through careful selective breeding. We will start an exercise today to see if you can accomplish the same thing.

How artificial selection differs from natural selection

In contrast to natural selection, artificial selection 1) favors traits that for some reason are favored by humans; 2) has a goal or direction toward which the selection process is directed; 3) generally is much faster than natural selection because the next generation can be absolutely restricted to offspring of parents that meet the desired criteria (natural selection is rarely so absolute). In artificial selection, humans are doing the selecting, intentionally restricting breeding to individuals with certain characteristics. In natural selection, the environment does the selecting – individuals that survive and reproduce better in a given environment, are “naturally selected”. The environment can include a large number of factors such as predators, food supply, weather, to name just a few.

Artificial Selection in Brassica, Part I Page 2 III. Experimental Procedure

A. Screening of First Generation of Plants

You and the other members of your lab section will participate as plant breeders in an effort to artificially select for a particular variable trait in a lineage of rapid-cycling Brassica rapa (Wisconsin Fast Plants™). The experiment will begin today but will not be completed for several weeks. The Brassica rapa variety you will be using is a product of intense artificial selection over the past 20 years for the following traits: rapid flowering and maturation; high seed production; short stature; ability to thrive under artificial light. The result of these efforts (conducted at the University of Wisconsin) is a valuable research and educational tool. The generation time (from seed to maturation to fertilization to mature seed) is 6-7 weeks, short enough to allow us to study one complete generation this semester. The life cycle of these so-called Wisconsin fast-plants is shown below.

Figure 2. Life cycle of rapid-cycling Brassica rapa.

Artificial Selection in Brassica, Part I Page 3 Although six weeks is remarkably short for a plant, this generation time is still slow for both research and educational purposes. We will need to rely on computer simulations for some of our other studies of genetics and evolution (such as the lab on the genetics of cat coat color and patterning).

We established our experimental population of Brassica rapa by planting commercially obtained seeds approximately 20 days prior to the lab. Thus, as you begin this study plants will have begun flowering (see Figure 2 above). These plants represent the initial population (the first generation). Your challenge as a class is to select, from this initial population, the top ten per cent of the plants that exhibit an extreme form of a specific variable trait, and to use this ‘selected’ lineage as parents of the succeeding generation. If successful, you will have demonstrated artificial selection, resulting in evolution within this particular lineage from one generation to the next.

Variable traits

What trait will you attempt to artificially select? There are a number of fairly obvious variable traits that one can observe in a large population of mature Fast Plants. A brief list might include: total number of flowers, total number of leaves, length of the lower leaf, surface area of the lower leaf, plant height, number of seeds per plant, total mass of plant, stem length between first and second leaves. In contrast, other traits usually don’t vary at all: number of petals per flower (four), petal color (bright yellow) and number of cotyledons or seed leaves (two).

Your lab instructor will give each student a container with ten plants. Treat these plants gently; they are young and tender, and easily damaged! Examine your plants and note the variability you can see. Remember, your plants are all the same age so differences you see (such as height or leaf number) are not due to differences in age.

List five traits that are variable within the small sample of plants you have at your table and give an indication of the degree of variability by noting the range (lowest and highest values).

1.

2.

3.

4.

5.

One variable that you might not have noted in your list is “hairiness” of leaves, petioles (leaf stalks) and stems but if you look more closely (especially with a hand lens), you should see these “hairs” or trichomes. Trichomes have been shown to have a specific function. In the space below, hypothesize what this function might be.

What is the function of trichomes?

Artificial Selection in Brassica, Part I Page 4 Counting “hairs”

As you have guessed, the variable trait you are going to try to alter in this lineage is “hairiness”. In the “wild type” population, some plants should be noticeably hairy, many are slightly hairy, while others are apparently hairless. Such an observation is too qualitative for our purposes. We need to a way to quantify hairiness.

To quantify hairiness, you could count all the trichomes on all parts of the plant. This task would be quite time-consuming and unnecessary. We know that the hairiness of one part of a plant is strongly correlated with hairiness on other parts. In other words, a plant’s hairiness in general can be quantified by assessing hairiness of a specific structure.

The structure we will use in this exercise is the petiole, or leaf stalk, where the trichomes are large, conspicuous and easily counted. The structure is relatively small with an easily defined starting and ending point.

To be consistent, we will use the petiole of the first (lowermost) true leaf and we will define the limits of the petiole as follows: from its junction with the main stem (usually marked by a small bulge or ridge, often differently colored to its junction with the lowermost leaf vein. See Figure 3 below.

Figure 3. Young Brassica rapa showing first true leaf, petiole and trichomes.

Note that the two lowermost, squarish, two-lobed and rather thick leaves are actually cotyledons (seed- leaves) and not true leaves. The first true leaf is just above the cotyledons. You now need to count the total number of trichomes on the petiole of the first true leaf of each plant in your sample. Your data will then be combined with data from the other students in this laboratory section.

Artificial Selection in Brassica, Part I Page 5 Use a hand lens and desk lamp. The trichomes are most conspicuous if strongly illuminated against a dark background, such as the black surface of the lab table. If present, the trichomes will generally be concentrated on the lower side of the petiole, but some can occur on the top or sides as well. Count a second time for verification and record your data in the spaces below.

Use lab tape and indelible marker to record the number of trichomes for the plants in each cell of your planter.

Enter your raw data in the data box below, and determine the median number of trichomes per petiole for your sample. The median is the middle value in a distribution, above and below which lie an equal number of values. Enter your data into the computer spreadsheet for the class.

Enter the number of trichomes on the petiole of the first true leaf, per plant: Median:

______

Combined data for original population

When all students have finished quantifying the trichome number of individual plants and entering data into the class spreadsheet, your lab instructor will use statistical software to generate a frequency histogram of the class data. Your instructor will make the summary available to you and you should copy the data into the frequency table below. Next, plot a frequency histogram of these data using the axes provided below.

Frequency Table for Class Data No. Of Trichomes Per Petiole Category for # of # of Trichomes Individuals 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 ≥50 Total # of Individuals # of Petiole Trichomes Median Trichome #

Artificial Selection in Brassica, Part I Page 6 B. Selecting the Parents of the Next (Second) Generation

Only a small fraction (about 10%) of this original population of plants will be selected to be parents of the next generation. These will not be randomly chosen, but instead will be the hairiest plants in the population. As a class, we need to identify the 14-18 plants that had the highest petiole trichome counts by referring to the label you created for each plant’s container. Plants other than the hairiest ones will be reserved for a subsequent study on inbreeding depression.

Record the trichome values of the selected individuals (parents-to-be) in the data box below.

Data Box

Selected parents (number of trichomes on petiole of first true leaf)

Selected individuals’ median value

Comparison of selected parents-to-be to original population

In the data box below, record the medians and their difference for the selected individuals and for the original population.

Data Box

Selected individuals, median

Original population, median

Difference

Artificial Selection in Brassica, Part I Page 7 C. Pollination of Selected Parental Plants

Brassica rapa plants need assistance in sexual reproduction because in nature they are totally dependent on certain insects for transferring sperm-bearing pollen from the male part of the flowers of one plant to the female part of the flowers of another plant. In these plants, the most conspicuous floral structures are the yellow petals. These petals enclose the male sexual structures (stamens, terminating in the pollen-producing anthers) and female structures (pistil, with the pollen-receiving stigma, style and egg- producing ovary), as shown in Figure 4 below.

Figure 4. Cross-section of flower of Brassica rapa.

Although each flower has both male and female parts, sperm (pollen) from one plant usually do not fertilize the eggs for the same plant. This self-incompatibility ensures that out-crossing (mating between different individuals) normally occurs.

The insect pollinators do not transfer pollen out of courtesy. Rather, the insects are lured to the flowers by the reward of nectar and edible pollen. The insects inadvertently pick up the sticky pollen on various body parts and then carry it with them to the next flower. Honeybees are common pollinators of Brassica in the field, and we will use honeybees (dead one) to help us pollinate the plants.

Artificial Selection in Brassica, Part I Page 8

Figure 5. Pollen on a bee. “Bee-sticks” have been prepared for you from the thoraxes of dead honeybees (collected from beekeepers after the bees died naturally – worker bees are short-lived). Glued to the end of a toothpick, these bee thoraxes make quite efficient pollinating devices; pollen grains cling to the many fine hairs on the bee’s body and are easily transferred to the stigmas of other flowers.

Performing pollination (ala group sex)

Each student should obtain one of the selected ‘parental’ plants. Your objective is to transfer pollen from each plant to every other plant at your table. This process can be done in the following way. Using a single bee-stick, each student will lightly rub and twirl the bee end for several seconds on the anthers and stigma of each open flower of her/his plant. Each then passes the bee-stick, now loaded with yellow pollen, to the neighboring student to the right, who will repeat the process. After each bee stick has made a complete round of the table (all open flower on all plants have been “visited” by all bee-sticks), the pollination process is completed. When finished discard the used bee-sticks; do not return the bee-sticks to their original container where other students might confuse them with fresh, unused ones.

Fertilization will result in the development of the ovules (each containing an embryo) into mature seeds, which will be contained in the fruit or seedpod (the elongated ovary). The length of time from fertilization to mature seed is about 3-4 weeks, sometimes a little longer. In subsequent weeks we will prune away the apical meristem plus additional flowers and axillary buds that may form in order to concentrate plant resources into the developing fruits and to hasten the seed-maturation process. Next week, you should easily be able to see the elongating fruits. In two weeks, they will have become even longer (2-5 cm) and the individual seeds inside (perhaps 5-20 per pod) will have become visible. In three weeks, the maturing process will be nearly complete. You will harvest and plant these seeds (the second generation) in five weeks.

Continuing pollination of new flowers as they develop during the week

To insure that enough viable seed will be produced from the selected ‘parental’ plants, the cross- pollination procedure will need to be continued for one week. Your lab instructor will devise a lab pollination schedule for which you will sign up to come in once during the week to pollinate the plants for your lab section. Your lab section’s plants will be grown in the Arey greenhouse.

Artificial Selection in Brassica, Part I Page 9 IV. Making Predictions About Hairiness of Second Generation

Now it is time to develop a hypothesis about the outcome of our experiment. How hairy do you think individuals in the second generation will be? Consider the three outcomes shown below. These are merely possibilities – none of them is necessarily “right” and there are many others not shown. Discuss these and others with your lab partner, then draw a frequency histogram in the blank graph at the bottom right of the Figure 4 below that you predict will represent the second generation (which will be about 100 offspring of the selected parents). Your histogram should show the range of values (lowest and highest) as well as the frequency (number of individuals that occur within each trichome number interval). Include on this histogram your prediction of the median value. The y-axis in your “predicted” histogram is intentionally unlabelled; you should fill in the interval values (in increments of either 5 or 10) so they are appropriate for your predictions.

Artificial Selection in Brassica, Part I Page 10